Search Program
Organizations
Contributors
Presentations
Production Session
Arts & Design
New Technologies
Production & Animation
Animation
Art
Artificial Intelligence/Machine Learning
Generative AI
Full Conference Supporter
Full Conference
DescriptionA young woman named Ada, desperate for a good night’s sleep, is kept awake by her exceedingly noisy neighbors. As she struggles to imagine what could be causing the cacophony upstairs, reality drifts into fantasy, and an epic battle for peace and sanity ensues.
That’s the story of “Dear Upstairs Neighbors”, an animated short made by a hybrid team of animation veterans and AI researchers.
The artistic core of the film — storyboards, concept art, character animation — was lovingly hand-crafted by artists using traditional 2D and 3D techniques. Our technical artists then created custom generative image and video models, which they fine-tuned on our own concept art and used, much like a renderer, to unite these hand-crafted components into polished shots that respect all of the artistic choices made along the way.
The film’s expressionistic painterly style evolves, shot by shot, to reflect Ada’s intense and changing emotions. To achieve this evolving look by hand would have meant painting over 7,000 frames, with artists spending so much effort on execution that they would have no time to iterate beyond a first pass. To do it using computer graphics would have required developing dozens of custom new stylized rendering algorithms — a feature film’s worth of R&D for a 6-minute short. Not impossible, but far from practical!
Instead, we built artist-driven, AI-assisted video-to-video workflows that let us iteratively refine the look of each shot, homing in on the exact stylistic choices that best suited each moment in the story. That process was itself a direct conversation between artists: our technical artists would render shots, and receive feedback from our director and production designer, often in the form of new paintings, which they would use to improve the fine-tuning for the next iteration.
This video-to-video approach gave us unprecedented control and flexibility. Each artist could work in their comfort zone, using whatever tool was most natural and creative for them. We could also iterate on every shot, switching freely between our generative models and traditional tools like Premiere and After Effects, to bring the shot closer to perfection. Masked generation allowed us to specify exactly where we wanted the models to improvise.
We feel that generative tools can only be truly useful for making animated films when artists are in control, not only as users of the models, but as co-designers. This film was our opportunity to take the wheel, and guide the technology’s development in a more artist-centered direction.
That’s the story of “Dear Upstairs Neighbors”, an animated short made by a hybrid team of animation veterans and AI researchers.
The artistic core of the film — storyboards, concept art, character animation — was lovingly hand-crafted by artists using traditional 2D and 3D techniques. Our technical artists then created custom generative image and video models, which they fine-tuned on our own concept art and used, much like a renderer, to unite these hand-crafted components into polished shots that respect all of the artistic choices made along the way.
The film’s expressionistic painterly style evolves, shot by shot, to reflect Ada’s intense and changing emotions. To achieve this evolving look by hand would have meant painting over 7,000 frames, with artists spending so much effort on execution that they would have no time to iterate beyond a first pass. To do it using computer graphics would have required developing dozens of custom new stylized rendering algorithms — a feature film’s worth of R&D for a 6-minute short. Not impossible, but far from practical!
Instead, we built artist-driven, AI-assisted video-to-video workflows that let us iteratively refine the look of each shot, homing in on the exact stylistic choices that best suited each moment in the story. That process was itself a direct conversation between artists: our technical artists would render shots, and receive feedback from our director and production designer, often in the form of new paintings, which they would use to improve the fine-tuning for the next iteration.
This video-to-video approach gave us unprecedented control and flexibility. Each artist could work in their comfort zone, using whatever tool was most natural and creative for them. We could also iterate on every shot, switching freely between our generative models and traditional tools like Premiere and After Effects, to bring the shot closer to perfection. Masked generation allowed us to specify exactly where we wanted the models to improvise.
We feel that generative tools can only be truly useful for making animated films when artists are in control, not only as users of the models, but as co-designers. This film was our opportunity to take the wheel, and guide the technology’s development in a more artist-centered direction.
Art Gallery
Arts & Design
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Full Conference Supporter
Full Conference
Experience
DescriptionSENSHO is a kinetic sculpture born from a project that applies 2D visual effects to physical objects. Referencing the Buddhist Vajra and rooted in the artist's filmmaking background, the work synchronizes high-speed strobes with mechanical vibration to generate real-time optical errors—chromatic aberration and glitches—on a solid object. Existing in the "In-Between" of 2D imagery and 3D matter, this visual phantom propagates to the body the moment it is touched, dissolving the boundary between the virtual image and the physical body.
Games Summit
Arts & Design
Gaming & Interactive
Art
Games
Performance
Real-Time
Rendering
Simulation
Full Conference Supporter
Full Conference
Experience
DescriptionBattlefield 6 redefines destruction as a tactical ecosystem rather than visual spectacle. Through ten interrelated destruction elements, systemic and pre-simulated workflows, consistent material damage classes, and shared pipelines, destruction reshapes gameplay spaces, enhances player agency, and delivers immersive, readable, and scalable battlefield experiences that fulfills franchaise vision.
Educator's Forum
Arts & Design
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Animation
Art
Augmented Reality
Capture/Scanning
Education
Ethics and Society
Generative AI
Math Foundations and Theory
Modeling
Performance
Pipeline Tools and Work
Real-Time
Rendering
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
Description“Live from LA!” is an immersive youth-theater production blending dance, poetry, and projection-mapping. We discuss how we scaffolded this production across disciplines, locations, and age groups to support youth political storytelling through personal narratives. We highlight how p5.js tools supported youth authorship while navigating practical constraints and safety.
Computer Animation Festival
Description“18 Months” is a true story about the unexpected paths to becoming a family. When Danny and Pete find a newborn in a subway station, their journey to adopt him is retold to mirror the stages of pregnancy—challenging traditional definitions and proving it doesn't always take a man, a woman, and 9 months to build a family.
Technical Paper
Research & Education
Geometry
Hardware
Modeling
Real-Time
Rendering
Full Conference Supporter
Full Conference
DescriptionWe introduce a state-of-the-art vectorization method using depth and semantic features to extract strokes. By modeling Bézier curves via differentiable 2D Gaussian splatting, we efficiently optimize geometry and texture. The framework extends to video for temporal consistency, offering high-quality, user-adjustable results for image and animation reconstruction.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Real-Time
Rendering
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Spatial Storytelling
Arts & Design
Gaming & Interactive
New Technologies
Art
Games
Generative AI
Lighting
Pipeline Tools and Work
Real-Time
Rendering
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
Description‘The Shadow of the Czar’ is the jazz-infused, Webby-honored noir game by James C. Kane featuring breakthrough use of 3D gaussian splat generation, reconstruction and rendering techniques in the browser. This demo will show how the hit cross-platform web title was developed solo and open sourced in just six weeks.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Geometry
Modeling
Rendering
Simulation
Full Conference Supporter
Full Conference
DescriptionWe reconstruct dynamic 4D scenes with multiple people from just 4 sparse cameras with minimal overlap. Our method separately handles backgrounds using video generation and humans using body model priors, then harmonizes the result with a temporally consistent enhancement step, achieving high-quality novel-viewpoint video from practical, everyday camera setups.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Geometry
Modeling
Rendering
Simulation
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Modeling
Rendering
Simulation
Full Conference Supporter
Full Conference
DescriptionWe pre-bake complex light scattering in 3D assets into neural 8D light transport functions by learning distributions from forward path-traced samples. Unlike previous far-field methods, our representation remains accurate under near-field illumination, while delivering lower variance and faster inference across challenging scenes than path tracing.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Generative AI
Geometry
Modeling
Rendering
Simulation
Full Conference Supporter
Full Conference
DescriptionWe introduce a Bayesian framework for task-specific next-best-view selection in 3D reconstruction. Unlike general uncertainty-reduction methods, we selectively reduce uncertainty only in regions critical to the downstream task via task-specific acquisition functions on posterior distributions over implicit surfaces, achieving superior performance with fewer views across classification, segmentation, and physics simulations.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Full Conference Supporter
Full Conference
DescriptionWe propose a unified few-step generative framework based on cumulative flow maps for long-range transport in probability space. By connecting instantaneous updates with finite-time transport, it applies broadly to diffusion and flow models, supports few-step and one-step generation, requires minimal modifications, preserves quality, and reduces inference cost across diverse tasks.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Modeling
Rendering
Simulation
Full Conference Supporter
Full Conference
DescriptionWe introduce a scalable 3D embedding that learns global illumination directly from 3D scene configurations. This design scales to environments with millions of triangles, enabling the first generalizable GI learning on complex, high-fidelity indoor scenes, far beyond prior limits. It bypasses per-scene retraining and screen-space artifacts to deliver realistic rendering.
Talks
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Animation
Dynamics
Games
Generative AI
Geometry
Modeling
Performance
Pipeline Tools and Work
Real-Time
Simulation
Full Conference Supporter
Full Conference
DescriptionRecent breakthroughs in generative modeling enable scaling 3D motion via sparse handles. We introduce the Generative Motion Rig (GMR), a Blender plugin bridging research and production. Through "generative keyframing," artists rapidly author or edit sequences using noise sampling and pose manipulation, streamlining the transition from traditional animation to AI-driven workflows.
Talks
Arts & Design
Production & Animation
Research & Education
Animation
Augmented Reality
Dynamics
Games
Geometry
Lighting
Math Foundations and Theory
Modeling
Pipeline Tools and Work
Real-Time
Rendering
Simulation
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionChoosing a BVH structure for accelerating ray tracing applications typically involves trading off between BVH build/update speed and BVH traversal speed. We introduce a hybrid BVH structure that combines the strengths of both the standard two-level structure and a monolithic single-level BVH, while minimizing the downsides of either approach.
Talks
Arts & Design
New Technologies
Production & Animation
Research & Education
Art
Artificial Intelligence/Machine Learning
Games
Generative AI
Industry Insight
Lighting
Modeling
Physical AI
Pipeline Tools and Work
Rendering
Scientific Visualization
Simulation
Full Conference Supporter
Full Conference
DescriptionWe present a scalable hybrid rendering system that addresses the scarcity of rare training data for autonomous vehicles by automatically compositing radiometrically calibrated synthetic objects into real-world HDR image.
Computer Animation Festival
Full Conference Supporter
Full Conference
DescriptionIn a quiet library tucked between memory and imagination, a child meets their future self. Through a poetic dialogue about fantasy, longing, and the beauty of growing up, they revisit the wonder of childhood stories and the dreams that shaped them. A Journey Through Pages is a reflective, hand-drawn short film that gently explores the bittersweet magic of moving forward without letting go of the worlds we once believed in.
Technical Paper
A LoD of Gaussians: Out-of-Core Training and Rendering for Seamless Ultra-Large Scene Reconstruction
2:40pm - 2:50pm PDT Monday, 20 July 2026 Room 408 AResearch & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Real-Time
Rendering
Full Conference Supporter
Full Conference
DescriptionGaussian Splatting struggles with large scenes due to memory limits and chunking artifacts. We introduce "A LoD of Gaussians," a new framework enabling ultra-large scene training and rendering on a single consumer GPU. Through out-of-core streaming and Level-of-Detail, it achieves seamless multi-scale reconstruction without relying on scene partitioning.
Emerging Technologies
Arts & Design
Gaming & Interactive
New Technologies
Research & Education
Art
Artificial Intelligence/Machine Learning
Audio
Augmented Reality
Capture/Scanning
Computer Vision
Digital Twins
Display
Dynamics
Education
Fabrication
Games
Generative AI
Graphics Systems Architecture
Haptics
Hardware
Image Processing
Performance
Physical AI
Pipeline Tools and Work
Real-Time
Robotics
Scientific Visualization
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionThe system is a MIDI-controlled interface that enables musical performance using water droplet collision sounds. It integrates solenoid valves for precise droplet timing and diameter control with tubular resonators for frequency-based pitch modulation. This setup achieves expressive rhythm, volume, and pitch through authentic acoustic sounds generated by actual physical phenomena.
Technical Paper
Research & Education
Animation
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionWe propose a unified nonlinear optimization framework that achieves a monolithic coupling of incompressibility, viscosity, and surface tension within a single solver. This first particle-based solver resolves their interdependence via position-based nonlocal viscosity formulations, enhancing stability beyond operator-splitting methods.
Technical Paper
Research & Education
Fabrication
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionThis paper presents a robust and efficient method for computing NURBS surface intersection curves, with emphasis on small loops and degenerate cases. By combining winding number theory with subdivision on the parametric domain, the method reliably detects starting points of all branches, improving topological correctness and computational efficiency in CAD.
Talks
Production & Animation
Animation
Modeling
Pipeline Tools and Work
Simulation
Full Conference Supporter
Full Conference
DescriptionUsdSkelPointInstancer is a new USD schema for crowds that combines packed arrays for agent data with prototypes built from shareable UsdSkel components. The increased speed and reduced memory of the new representation was used successfully to represent large crowds on Walt Disney Animation Studios' "Zootopia 2."
Talks
Arts & Design
Production & Animation
Research & Education
Animation
Augmented Reality
Dynamics
Games
Geometry
Lighting
Math Foundations and Theory
Modeling
Pipeline Tools and Work
Real-Time
Rendering
Simulation
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionWalt Disney Animation Studios' "Versa" is a beautifully emotional six minute short, where volumes make up everything across both environments and characters. This talk explores how we enabled new workflows and achieved cross-departmental collaboration with Effects to bring this unique short to life.
Course
Gaming & Interactive
Production & Animation
Research & Education
Graphics Systems Architecture
Rendering
Scientific Visualization
Full Conference Supporter
Full Conference
DescriptionThis course introduces the Khronos ANARI rendering API using a highly practical, step-by-step approach in which attendees learn to write increasingly capable ANARI based applications. At the end of this course, users should be easily able to realize their own 3D applications using ANARI for both interactive and offline rendering use cases.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Augmented Reality
Rendering
Simulation
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionWe present a nearly perspective-correct video passthrough headset with just 2 milliseconds of end-to-end latency. We use this headset to study perceptual requirements for low-latency passthrough in head-mounted displays.
Technical Paper
Research & Education
Fabrication
Geometry
Modeling
Robotics
Simulation
Full Conference Supporter
Full Conference
DescriptionWe present a unified computational framework for numerical homogenisation of straight- and curved-crease origami materials with periodic crease patterns.
Art Gallery
Arts & Design
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Full Conference Supporter
Full Conference
Experience
DescriptionPlease visit our website: https://www.zhangweidi.com/awalledcity
Password for Private Documentation Video: AM3
A Walled City is an interactive, generative monument to memory, density, and collective digital presence. Grounded in the spatial logic, zoning, and aesthetics of Hong Kong’s historical Kowloon Walled City, a self-built, hyper dense, decentralized enclave that grew within a political void from the 1950s to 1993, the artist created a customized multi-agent AI system that transforms participant uploaded images into unique architectural chambers. Each chamber functions as a mnemonic capsule that assembles into an evolving urban interface for non hierarchical collective speculation.
Password for Private Documentation Video: AM3
A Walled City is an interactive, generative monument to memory, density, and collective digital presence. Grounded in the spatial logic, zoning, and aesthetics of Hong Kong’s historical Kowloon Walled City, a self-built, hyper dense, decentralized enclave that grew within a political void from the 1950s to 1993, the artist created a customized multi-agent AI system that transforms participant uploaded images into unique architectural chambers. Each chamber functions as a mnemonic capsule that assembles into an evolving urban interface for non hierarchical collective speculation.
Art Paper
Arts & Design
Animation
Art
Artificial Intelligence/Machine Learning
Capture/Scanning
Computer Vision
Generative AI
Image Processing
Modeling
Pipeline Tools and Work
Real-Time
Rendering
Full Conference Supporter
Full Conference
Experience
DescriptionIt proposes a new artistic model of collective virtual presence grounded in historical urbanism and decentralized networks. Drawing from the spatial logic and aesthetic influence of the Kowloon Walled City, the project provides a speculative approach to participatory spatial memory systems through interactive AI art experience.
Course
Arts & Design
Gaming & Interactive
Research & Education
Animation
Education
Lighting
Modeling
Rendering
Full Conference Supporter
Full Conference
Experience
DescriptionFor a beginner, walking into SIGGRAPH is a very intimidating experience. There is so much to see and so much to do, but everyone seems to be speaking an unfamiliar language. No matter who you are, this course is for you! It will give everyone a background in concepts needed to get more from the conference.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Fabrication
Generative AI
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionArtistic styles often involve shape reinterpretation, not merely a change of surface appearance. Conventional style transfer methods miss this deeper abstraction behavior. We introduce Abstraction in Style (AiS), a two-stage generative framework that learns and transfers style at shape level, beyond texture and colors, enabling more expressive and controllable stylization.
ACM SIGGRAPH Award Talks
Full Conference Supporter
Full Conference
Experience
DescriptionEach year, ACM SIGGRAPH presents nine awards recognizing exceptional achievements in computer graphics and interactive techniques at the ACM SIGGRAPH Conference.
For a list of the awardees, visit: https://www.siggraph.org/awards/
For a list of the awardees, visit: https://www.siggraph.org/awards/
ACM SIGGRAPH 365
Research & Education
Education
Scientific Visualization
Full Conference Supporter
Full Conference
Experience
DescriptionThe ACM SIGGRAPH Cartographic Visualization (Carto) session explores how viewpoints and techniques from the computer graphics community can be effectively applied to cartographic and spatial data sets. Speakers demonstrate their latest tools and application efforts.
ACM SIGGRAPH 365
Arts & Design
Gaming & Interactive
Research & Education
Art
Full Conference Supporter
Full Conference
Experience
DescriptionArt Community Roundtables for Art Papers, Art Gallery, and Digital Arts Community members. Community members will be invited to join conversations related to the Art Gallery, Art Papers, and Digital Arts Community activities, as well as other themes related to the arts at SIGGRAPH.
ACM SIGGRAPH 365
Research & Education
Industry Insight
DescriptionYearly Pioneers social gathering with light appetizers and a cash bar. No agenda, just schmoozing with old friends and colleagues. Admission by Pioneer membership as printed on your badge.
ACM SIGGRAPH 365
Arts & Design
Gaming & Interactive
Production & Animation
Research & Education
Art
Education
Full Conference Supporter
Full Conference
Experience
DescriptionThe kickoff of the 2026 SIGGRAPH Educators Program will include an overview of activities and opportunities with, and sponsored by, the ACM SIGGRAPH Education Committee. Additionally, the winning entries from the SpaceTime competition will be shown along with a screening of the show reel for the 2026 double-curated Faculty Submitted Student Work (FSSW) exhibit. Designed as a way for educators to share their project ideas across schools and disciplines, FSSW is an online archive of assignment and project briefs as well as a curated video emphasizing the variety of student work and schools submitted. The annual exhibition video is a taste of the content available on the Education Committee website and this event showcases and celebrates to the greater SIGGRAPH community the best examples of student work done for projects and assignments for the 2025-2026 school year. Part 2 of this event will begin immediately upon the conclusion of presentations in Part 1.
ACM SIGGRAPH 365
Arts & Design
Gaming & Interactive
Production & Animation
Research & Education
Art
Education
Full Conference Supporter
Full Conference
Experience
DescriptionThis session is the second part of the kickoff for the 2026 SIGGRAPH Educators Program and will primarily be the screening of the annual show reel for the 2026 double-curated Faculty Submitted Student Work (FSSW) exhibit. Designed as a way for educators to share their project ideas across schools and disciplines, FSSW is an online archive of assignment and project briefs as well as a curated video emphasizing the variety of student work and schools submitted. The annual exhibition video is a taste of the content available on the Education Committee website and this event showcases and celebrates to the greater SIGGRAPH community the best examples of student work done for projects and assignments for the 2025-2026 school year. NOTE: Part 2 will begin immediately upon completion of other presentations by the ACM SIGGRAPH Education Committee that begins in Part 1.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionACT is a unified AI framework that effortlessly brings static 3D characters to life. By integrating rigging, skinning, and motion generation into a single diffusion-based pipeline, ACT automatically animates characters of any shape, delivering high-quality, physically plausible motions that overcome the limitations of traditional, multi-step animation workflows.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Generative AI
Rendering
Full Conference Supporter
Full Conference
DescriptionActCam is a zero-shot method providing joint control over camera trajectories and character acting. It integrates into preexisting artistic workflows by leveraging traditional cinematography skills for fine-grained control. Requiring no additional training, the framework avoids costly finetuning and adapts across different backbone architectures.
Technical Paper
Research & Education
Display
Generative AI
Hardware
Image Processing
Full Conference Supporter
Full Conference
DescriptionThis work provides a recipe for adapting existing image/video quality metrics for evaluating tone-mapping algorithms. We demonstrate across multiple datasets that when tone-mapped images and their HDR reference are represented in a perceptually-uniform space, existing metrics such as TOPIQ, DISTS, and ColorVideoVDP outperform tone-mapping-specific metrics, such as TMQI and FFTMI.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Robotics
Simulation
Full Conference Supporter
Full Conference
DescriptionWe propose a motion in-betweening method tailored to stylized keyframe-based animation production. Our Adaptive Interpolation–Synthesis layer dynamically blends learned interpolation with direct pose synthesis, mirroring the animator's workflow. Combined with a domain-based input keypose training schedule, our approach achieves state-of-the-art results and accelerates in-betweening by 3.5× in production workflows.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Lighting
Modeling
Real-Time
Rendering
Full Conference Supporter
Full Conference
DescriptionThis paper accelerates Gaussian Process Implicit Surface (GPIS) rendering by introducing an online algorithm for incrementally sampling conditional Gaussian distributions along rays and an adaptive ray marching scheme that minimizes steps. Together, these techniques reduce rendering error by up to 46× at equal time compared to prior methods.
Computer Animation Festival
DescriptionAddicted To It follows Dace, who attempts the impossible: a peaceful park walk without his phone. Unfortunately, his phone—Phoney Bunny—has other plans and literally springs to life to win him back. Dace soldiers on, recruiting fellow screen junkies to unplug, while Phoney Bunny chases him like a clingy ex with 100% battery. It’s a fast, ridiculous visual feast with zero solutions—just one simple question: can you survive a 5–10 minute walk without your phone?
Technical Paper
Research & Education
Fabrication
Geometry
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionWe present ADS, a method that delivers both pseudo-random surface samples and isosurface connectivity from occupancy functions, unlike prior approaches that typically provide only one. ADS achieves this with orders of magnitude fewer function evaluations at comparable accuracy than prior approaches, by progressively refining a Delaunay scaffold around the surface.
Course
Production & Animation
Research & Education
Performance
Rendering
Scientific Visualization
Full Conference Supporter
Full Conference
DescriptionThis course teaches attendees how to write sci-vis volume renderers on modern architectures. Since the last course at Eurographics 2006 the hardware and algorithmic landscape have changed tremendously. We show how this has influenced the state-of-the-art in real-time volume graphics in the last years.
Course
Gaming & Interactive
Artificial Intelligence/Machine Learning
Games
Rendering
Full Conference Supporter
Full Conference
DescriptionModern video games employ a variety of sophisticated algorithms to produce groundbreaking 3D rendering pushing the visual boundaries and interactive experience of rich environments. This course brings state-of-the-art and production-proven rendering techniques for fast, interactive rendering of complex and engaging virtual worlds of video games.
This is the course to attend if you are in the game development industry or want to learn the latest and greatest techniques in the real-time rendering domain!
This is the course to attend if you are in the game development industry or want to learn the latest and greatest techniques in the real-time rendering domain!
Course
Gaming & Interactive
Artificial Intelligence/Machine Learning
Games
Rendering
Full Conference Supporter
Full Conference
DescriptionModern video games employ a variety of sophisticated algorithms to produce groundbreaking 3D rendering pushing the visual boundaries and interactive experience of rich environments. This course brings state-of-the-art and production-proven rendering techniques for fast, interactive rendering of complex and engaging virtual worlds of video games.
This is the course to attend if you are in the game development industry or want to learn the latest and greatest techniques in the real-time rendering domain!
This is the course to attend if you are in the game development industry or want to learn the latest and greatest techniques in the real-time rendering domain!
Talks
Production & Animation
Research & Education
Animation
Art
Capture/Scanning
Digital Twins
Dynamics
Industry Insight
Lighting
Pipeline Tools and Work
Rendering
Simulation
Full Conference Supporter
Full Conference
DescriptionIn this talk we discuss the collection of improvements we have made to our in-house water system Pahi to deliver the unprecedented scale and complexity of Avatar: Fire and Ash.
Panel
Production & Animation
Artificial Intelligence/Machine Learning
Computer Vision
Industry Insight
Lighting
Pipeline Tools and Work
Real-Time
Full Conference Supporter
Full Conference
DescriptionExploring recent advances in virtual production resulting in higher performance, better data interchange, and realtime VP scalability for VFX supervisors and filmmakers. From lumenAOVs to frame generation with nDisplay and fast rendering Gaussian Splats, speakers discuss the integration of Epic and NVIDIA technologies, technical opportunities, pipeline impacts, and collaborative workflows.
Technical Paper
Research & Education
Animation
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Generative AI
Geometry
Modeling
Simulation
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionAGILE reconstructs 3D hand-object interactions from monocular videos via agentic generation. It produces complete, simulation-ready object meshes and robustly estimates object pose without SfM, enabling accurate and stable results on challenging real-world data.
Technical Paper
Research & Education
Animation
Geometry
Hardware
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionWe propose a GPU-friendly adaptive algebraic coarsening method guided by Green strain increments and combined with affine embedding to avoid topological changes and irregular memory access, yielding the first fully GPU-optimized adaptive IPC solver with up to 3× speedup over state-of-the-art GPU IPC while preserving identical visuals.
Art Paper
Arts & Design
Animation
Art
Artificial Intelligence/Machine Learning
Audio
Capture/Scanning
Computer Vision
Display
Ethics and Society
Generative AI
Hardware
Image Processing
Physical AI
Real-Time
Robotics
Scientific Visualization
Simulation
Spatial Computing
Full Conference Supporter
Full Conference
Experience
DescriptionAfter all the presentations, attendees are invited to participate in a Q&A.
Course
Arts & Design
Production & Animation
Art
Computer Vision
Generative AI
Full Conference Supporter
Full Conference
DescriptionThis course introduces a multi-agent framework fusing Generative AI’s creativity with Computer Graphics’ precision. Addressing GenAI’s limitations in data accuracy, we demonstrate how agents orchestrate deterministic rendering (e.g., for charts or spatial logic) to guide video synthesis. Attendees will learn to build systems for controllable, precise cinematic and data storytelling.
Technical Workshop
Arts & Design
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Art
Artificial Intelligence/Machine Learning
Computer Vision
Generative AI
Image Processing
Full Conference Supporter
Full Conference
DescriptionThe 10th installment of the CVEU workshop series marks a transformative shift toward the Next-Generation Creator-Machine Co-Learning Paradigm. Organized by leading academic researchers and the technical architects behind current video generation models, this workshop explores a future where generative AI evolves from a passive tool into a collaborative partner that internalizes both physical laws and artistic intent. Our primary objective is to bridge the "semantic gap" between raw visual tokens and the sophisticated language of cinema, focusing on how World Models can balance physical factuality with creative license. We will investigate how machine learning translate abstract user intent into precise model behavior, effectively reconstructing traditional production pipelines for professional filmmakers and artists alike. Attendees can expect a high-energy program featuring technical demos of AI-native creative workflows, provocative keynote sessions, and an open "Research-meets-Art" roundtable designed to spark dialogue between model builders and visual storytellers. This forum offers a unique opportunity to engage directly with the creators of the world’s most advanced visual AI systems to define the next era of human-machine synergy in the digital arts.
Spatial Storytelling
Arts & Design
Research & Education
Art
Ethics and Society
Scientific Visualization
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionAI War Cloud Database catalogues the systems used to make automated decisions in warfare, and it maps where the same types of tools appear in smartphones and popular social platforms. This demonstration presents critical and technical decisions that were hand-in-hand, central to research, information design, and proliferation of the experience.
Talks
Arts & Design
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Animation
Art
Artificial Intelligence/Machine Learning
Ethics and Society
Generative AI
Industry Insight
Pipeline Tools and Work
Full Conference Supporter
Full Conference
DescriptionDramatic changes in audience behavior, platforms, and viewing habits combined with runaway production to create an existential crisis in Hollywood, even before the arrival of Cinematic AI. While AI is both an accelerant and a solution, the long-term reality is that the algorithm will transform media as we know it.
Immersive Pavilion
Arts & Design
Gaming & Interactive
New Technologies
Research & Education
Animation
Art
Artificial Intelligence/Machine Learning
Audio
Augmented Reality
Computer Vision
Digital Twins
Education
Ethics and Society
Fabrication
Games
Generative AI
Haptics
Hardware
Image Processing
Modeling
Performance
Pipeline Tools and Work
Real-Time
Scientific Visualization
Simulation
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionAs mixed reality becomes ubiquitous, we risk a fragmented future where users inhabit isolated worlds. TouchPort-Allow Me Into Your Dream is a multi-person sharing protocol that reprograms handshakes: when two MR wearers shake hands, a portal opens over their clasped hands, letting each peek into the other's world and “allowing” them into their “dreams”.
Technical Paper
Research & Education
Lighting
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionThis paper presents an ambient-robust inverse rendering method using robot-assisted RGB–NIR imaging. By combining multi-view RGB images with NIR flash images, we reconstruct accurate geometry and reflectance under multiple lighting conditions. We also introduce a robot-assisted imaging system and multi-view RGB–NIR dataset collected across multiple ambient illumination settings.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Generative AI
Geometry
Modeling
Simulation
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionExisting Human-Mesh-Recovery methods struggle with dynamic airborne motions, often producing unstable and unrealistic results. We present a refinement approach that enforces physical consistency by selecting reliable motion segments, applying homotopy-aware trajectory optimization. Experiments on in-the-wild and challenging videos show improved realism and stability over prior methods.
Talks
Arts & Design
Gaming & Interactive
Production & Animation
Research & Education
Art
Computer Vision
Digital Twins
Education
Games
Image Processing
Pipeline Tools and Work
Rendering
Full Conference Supporter
Full Conference
DescriptionTransmissive materials are difficult to parameterize in an artist-friendly way. This talk presents a practical albedo approximation derived from a simplified Kubelka–Munk model. Using thin- and thick-layer reflectance, absorption is recovered in closed form, avoiding iterative fitting and enabling efficient, intuitive control for real-time rendering applications.
Course
Gaming & Interactive
New Technologies
Research & Education
Artificial Intelligence/Machine Learning
Games
Math Foundations and Theory
Rendering
Full Conference Supporter
Full Conference
DescriptionLearn neural shading from the ground up. This hands-on course teaches attendees to build and train MLPs directly in Slang shaders, leverage GPU tensor cores for hardware acceleration, and implement neural materials — transforming complex visual phenomena into compact, high-fidelity learned representations.
Course
New Technologies
Research & Education
Artificial Intelligence/Machine Learning
Computer Vision
Games
Generative AI
Image Processing
Math Foundations and Theory
Simulation
Full Conference Supporter
Full Conference
DescriptionQuantum computing is a radically new and exciting approach to programming. By exploiting the unusual behavior of quantum objects, this new technology invites us to re-imagine the computer graphics methods we know and love in revolutionary new ways. This course is math-free and requires no technical background.
Course
Production & Animation
Research & Education
Audio
Computer Vision
Education
Image Processing
Math Foundations and Theory
Modeling
Rendering
Scientific Visualization
Full Conference Supporter
Full Conference
DescriptionThe Fourier Transform is fundamental to computer graphics, explaining topics from aliasing and sampling to image compression and filtering. This friendly course explains the principles in words, pictures, and animation, rather than math. The concepts are the important thing. We show that they are comprehensible, useful, and beautiful.
Spatial Storytelling
Arts & Design
Gaming & Interactive
Research & Education
Art
Ethics and Society
Fabrication
Real-Time
Robotics
Spatial Computing
Full Conference Supporter
Full Conference
Experience
DescriptionThis talk presents a place-based approach to designing interactive systems that combine ancestral craft and emerging technologies. Through XR installations and robotic artworks developed in the Amazon rainfores, it shares methods for building non-generic futures grounded in local knowledge, material practices, collaboration and real-time graphics.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Full Conference Supporter
Full Conference
DescriptionAniGen generates animatable 3D assets from a single image by jointly producing geometry, skeletons, and skinning weights. Its unified S³ field representation enables consistent, high-quality rigged asset creation across diverse categories, including animals, humans, cartoon characters, and articulated objects.
Educator's Forum
Arts & Design
Gaming & Interactive
Research & Education
Animation
Art
Education
Games
Scientific Visualization
Full Conference Supporter
Full Conference
Experience
DescriptionThis talk presents four project-based learning experiences from a small public university animation program, demonstrating how connecting technology, artistry, and purpose fosters student growth and engagement. Through classroom, collaborative, and community-based projects, the talk offers transferable strategies for supporting diverse student populations in evolving creative fields.
Talks
Arts & Design
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Art
Artificial Intelligence/Machine Learning
Capture/Scanning
Computer Vision
Games
Generative AI
Performance
Physical AI
Pipeline Tools and Work
Real-Time
Rendering
Robotics
Scientific Visualization
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionThis talk introduces a new public library, designed for rapid prototyping of interactive Web interfaces over a large class of emerging technologies in 3D representations and AI. We base our design on first-hand experience in developing multiple interactive tools presented at SIGGRAPH as technical papers, Real-Time Live presentations, and Labs.
Computer Animation Festival
DescriptionAmid the harsh realities of South African apartheid, two boys—Themba and Joel—form a dangerous friendship. As hate and fear threaten to pull them apart, their bond faces its greatest challenge in the form of a life-saving gesture. Through their eyes, we glimpse a moment of innocence—where hope, empathy, and understanding still feel possible.
Art Gallery
Arts & Design
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Full Conference Supporter
Full Conference
Experience
DescriptionAre We Gazing at the Same Moon? Do we observe things within the same time and space?
The artwork established a moon-gazing ritual that re-imagines the moon as a concurrent presence across parallel realities, questioning the relationship between personal memory and collective memory. Challenging the process of "Lunar Mosaic" by reversing the process and expanding image segments of the collaged moon with generative AI, it questions the notion of a singular, unified human experience of moon-gazing. The project spans various forms, including audio-visual works, sculpture, and public installation.
The artwork established a moon-gazing ritual that re-imagines the moon as a concurrent presence across parallel realities, questioning the relationship between personal memory and collective memory. Challenging the process of "Lunar Mosaic" by reversing the process and expanding image segments of the collaged moon with generative AI, it questions the notion of a singular, unified human experience of moon-gazing. The project spans various forms, including audio-visual works, sculpture, and public installation.
Educator's Forum
Gaming & Interactive
New Technologies
Research & Education
Augmented Reality
Digital Twins
Education
Games
Image Processing
Real-Time
Scientific Visualization
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionWidely used for making and teaching algorithmic art, p5.js prioritizes access and expressiveness. What impact does p5.js have on students? This presentation reflects on an educator program that grounds web-based interactive graphics in students' contexts, expanding imagination of what is possible with resources at hand, and suggesting new professional possibilities.
ACM SIGGRAPH 365
Art Gallery
Art Paper
Arts & Design
Research & Education
Art
Artificial Intelligence/Machine Learning
Generative AI
Full Conference Supporter
Full Conference
Experience
DescriptionWelcome to the Art Gallery and Art Papers Contributors from the Chairs, and announcement of the Best in Show awards for those programs. Acknowledgement of the Distinguished Artist Award winner, and recognition of the contributions of the year-round Digital Arts Community (DAC).
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Rendering
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionWe present a few-shot method for articulating static 3D objects using text-guided motion personalization. Leveraging diffusion models, we generate plausible part motions and adapt them to input objects via image prompting. Differentiable rendering with score distillation transfers multi-view motion into 3D, enabling optimization of articulation parameters under piecewise rigidity assumptions.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Generative AI
Geometry
Modeling
Rendering
Simulation
Full Conference Supporter
Full Conference
DescriptionArtiFixer enhances and extends 3D reconstructions efficiently and scalably. Given an initial reconstruction and reference views, it autoregressive generates novel content consistent with existing observations. ArtiFixer can produce hundreds of novel views in a single inference pass or serve as pseudo-supervision to improve the underlying 3D reconstruction.
Technical Workshop
New Technologies
Research & Education
Generative AI
Geometry
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionNeural graphics is transforming how we create, represent, and interact with visual content. This workshop brings together leading researchers across geometry, rendering, generative modeling, and embodied AI to present a unified view of this rapidly evolving field.Through a curated sequence of invited talks, the program will guide attendees from geometric foundations and 3D generation to neural rendering, real-time techniques, physics-aware reconstruction, and interaction with the physical world.
The primary objective is to explore how neural representations and generative models can be integrated with classical graphics pipelines to enable more scalable, controllable, and physically grounded visual computing. The workshop will feature the technical presentations, encouraging active engagement between speakers and attendees. Participants can expect lively exchanges, opportunities to ask questions, and insights into emerging research directions.
Designed for researchers, practitioners, and students in graphics, vision, and AI, this workshop offers a concise yet comprehensive overview of neural graphics and its future trajectory.
The primary objective is to explore how neural representations and generative models can be integrated with classical graphics pipelines to enable more scalable, controllable, and physically grounded visual computing. The workshop will feature the technical presentations, encouraging active engagement between speakers and attendees. Participants can expect lively exchanges, opportunities to ask questions, and insights into emerging research directions.
Designed for researchers, practitioners, and students in graphics, vision, and AI, this workshop offers a concise yet comprehensive overview of neural graphics and its future trajectory.
Talks
Production & Animation
Animation
Pipeline Tools and Work
Real-Time
Full Conference Supporter
Full Conference
DescriptionThis talk introduces Assets in Motions (AIM) at Sony Pictures Imageworks. AIM is a strategic approach, designed to move character designs into well rounded production ready assets in a more cost effective and collaborative way.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Geometry
Modeling
Rendering
Simulation
Full Conference Supporter
Full Conference
DescriptionATGS is a volumetric video reconstruction method that handles long sequences with complex motions and supports rendering from arbitrary viewpoints, enabling immersive experiences.
Technical Paper
Research & Education
Animation
Fabrication
Haptics
Simulation
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionAtmospheric Haptics treats airflow and temperature as programmable, non-contact tactile fields for VR. Four perceptual primitives—Intensity, Temperature, Spatiality, and Dynamics—drive a three-layer rendering pipeline implemented in a handheld prototype. A user study confirms enhanced realism, improved navigation precision, and convincing wetness illusions without liquid contact.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Fabrication
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionAtomSlicer is a 3D printing method for fused filament fabrication that generates non-planar layers and continuous toolpaths aligned with user-defined fields. It enables better control of layer orientation, constant thickness, and near-continuous deposition, helping improve print quality, reduce interruptions, and support advanced multi-axis fabrication.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Games
Generative AI
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionAudio-Omni is the first unified framework for multi-modal audio generation and editing across sound, music, and speech. By combining a frozen multimodal language model with a trainable diffusion transformer, it achieves state-of-the-art performance on multiple benchmarks while exhibiting remarkable inherited capabilities.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Full Conference Supporter
Full Conference
DescriptionParaCAD is a point-cloud-conditioned autoregressive framework for B-Rep generation that directly predicts native parametric surface types and continuous parameters via a surface-centric tokenization, preserving CAD semantics and downstream editability beyond grid/point-based representations.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Robotics
Simulation
Full Conference Supporter
Full Conference
DescriptionARDY is a real-time autoregressive diffusion model for interactive 3D human motion generation. Utilizing a novel hybrid motion representation, ARDY empowers users to interactively synthesize high-fidelity animations directed by dynamic text prompts and flexible spatial constraints, such as full-body keyframes, trajectories, and end-effector joints.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Generative AI
Lighting
Full Conference Supporter
Full Conference
DescriptionThis work introduces an autoregressive transformer designed for video montage, transforming raw footage into cinematic sequences. We train it on large corpora of movies to learn film grammar. Instead of generating pixels, we introduce a footage-constrained decoding algorithm to select shots from the input raw footage.
Talks
Arts & Design
Production & Animation
Animation
Art
Industry Insight
Lighting
Pipeline Tools and Work
Rendering
Full Conference Supporter
Full Conference
DescriptionFlying Bark Productions introduces an NPR workflow for Avatar Aang: The Last Airbender (feature) that shifts final stylization to Compositing, integrating directly with background paintings. Using Local Position and Tangent Field Vectors, it generates and smears procedural strokes in 2D screen space, enabling painterly stylization, dynamic lighting, and temporal stability.
Talks
Production & Animation
Animation
Pipeline Tools and Work
Real-Time
Full Conference Supporter
Full Conference
DescriptionParamount Animation formed a multidisciplinary “Strike Force” during Avatar Aang: The Last Airbender (feature) to support a distributed, vendor‑based pipeline. The team paralleled work across global studios while building an in‑house pipeline for finaling, FX, layout, and compositing. This document details the group’s purpose, workflows, tools, and outcomes under tight constraints.
Computer Animation Festival
Full Conference Supporter
Full Conference
DescriptionHadenna, a young warrior, travels across the galaxy in search of the ultimate cure to restore her brother, Theos, to his true body.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Full Conference Supporter
Full Conference
DescriptionCurrent language-guided CAD editing struggles with dataset shortages and relies on construction history. We introduce B-repLer, a novel framework enabling text-based CAD edits in a learned latent space without requiring construction history. We also present BrepEDIT-240K, the first large-scale editing dataset, enabling accurate, complex CAD edits from high-level text commands.
Educator's Forum
Arts & Design
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Animation
Art
Augmented Reality
Digital Twins
Education
Games
Generative AI
Real-Time
Simulation
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionBeaver World uses beaver-based simulation to teach watershed restoration, integrating Indigenous ecological knowledge with climate education.
Talks
Arts & Design
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Art
Artificial Intelligence/Machine Learning
Capture/Scanning
Computer Vision
Games
Generative AI
Performance
Physical AI
Pipeline Tools and Work
Real-Time
Rendering
Robotics
Scientific Visualization
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionReVerie: Sense is an interactive and co-creative fulldome installation that explores the use of 3D generative AI and a multisensory robotic sculpture for transmedia storytelling. In this talk, we present the creative R&D process, offering reflections on designing with generative AI for expanded storytelling with an immersive and somatic approach.
Spatial Storytelling
Arts & Design
New Technologies
Research & Education
Capture/Scanning
Digital Twins
Spatial Computing
Full Conference Supporter
Full Conference
Experience
Description"The New York Times has an R&D team?" In our industry dedicated R&D is rare. We make space for technologists and journalists to pilot new ways of reporting stories. Learn how we navigate journalistic constraints, pioneer spatial storytelling techniques, and bridge gaps between emerging technology and traditional journalism workflows.
ACM SIGGRAPH 365
Research & Education
Animation
Artificial Intelligence/Machine Learning
Audio
Augmented Reality
Capture/Scanning
Computer Vision
Digital Twins
Display
Dynamics
Education
Ethics and Society
Fabrication
Games
Generative AI
Geometry
Graphics Systems Architecture
Haptics
Hardware
Image Processing
Industry Insight
Lighting
Math Foundations and Theory
Modeling
Performance
Physical AI
Pipeline Tools and Work
Real-Time
Rendering
Robotics
Scientific Visualization
Simulation
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionFollowing the success of last year’s edition, this session will present a curated selection of outstanding research contributions from the Eurographics 2026 conference. The session aims to provide SIGGRAPH attendees with insights into current trends, emerging directions, and key developments in Computer Graphics and related fields, with a particular focus on research originating from the European community and its global impact.
The session will open with a short introduction to Eurographics, followed by presentations of the selected works and awards, in particular:
- The Best Paper Award of Eurographics 2026
- The Selected Honourable Mentions of the Best Paper Award of Eurographics 2026
- The recipient of the Eurographics Young Researcher Award 2026
The session will open with a short introduction to Eurographics, followed by presentations of the selected works and awards, in particular:
- The Best Paper Award of Eurographics 2026
- The Selected Honourable Mentions of the Best Paper Award of Eurographics 2026
- The recipient of the Eurographics Young Researcher Award 2026
Technical Paper
Research & Education
Animation
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionWhile thin shells have been studied for decades, there is little consensus on how to discretize them. We systematically study bending models for Kirchhoff-Love shells. We develop new models and methods to address discovered simulation gaps, and make practical recommendations of when and how these models should be used.
Course
New Technologies
Research & Education
Artificial Intelligence/Machine Learning
Capture/Scanning
Digital Twins
Generative AI
Geometry
Physical AI
Rendering
Scientific Visualization
Simulation
Full Conference Supporter
Full Conference
DescriptionDiscover how scalable compute, advanced graphics and generative AI reshape scientific and cultural narratives. Leveraging AI and neural graphics, we guide attendees through the complete object lifecycle - Seeing, Reconstructing, Restoring, and Simulating. Learn to build active, physically grounded digital twins that unlock deep insights across scientific and cultural domains.
ACM SIGGRAPH 365
Arts & Design
Ethics and Society
Generative AI
Full Conference Supporter
Full Conference
Experience
DescriptionComputational design reshaped interactive media, product design, and architecture over the past two decades. What comes next?
This SIGGRAPH 365 session puts that question to the ACM SIGGRAPH Design Affinity Group, tool-makers and educators who have shaped computational design practice and its platforms. In a sequence of short provocations, panelists will stake out what supersedes computational design: where AI, generative methods, and open frameworks are taking creative practice, and how future tools can support human agency and collaborative forms of design.
The talks will open a discussion on bridging domains: interaction and UX designers, visual and graphic designers, industrial and product designers, as well as architects, have been part of SIGGRAPH since its inception, but have remained at its edges. The Design Affinity Group exists to move them closer to the center. We want your take on what advocacy for design within SIGGRAPH should look like.
This SIGGRAPH 365 session puts that question to the ACM SIGGRAPH Design Affinity Group, tool-makers and educators who have shaped computational design practice and its platforms. In a sequence of short provocations, panelists will stake out what supersedes computational design: where AI, generative methods, and open frameworks are taking creative practice, and how future tools can support human agency and collaborative forms of design.
The talks will open a discussion on bridging domains: interaction and UX designers, visual and graphic designers, industrial and product designers, as well as architects, have been part of SIGGRAPH since its inception, but have remained at its edges. The Design Affinity Group exists to move them closer to the center. We want your take on what advocacy for design within SIGGRAPH should look like.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionThis paper introduces two new compact neural encodings, the hash-sphere and hash-grid-sphere, which represent all-frequency signals using hierarchical geodesic grids. While we demonstrated their effectiveness in neural path guiding, radiance field reconstruction, and incident radiance caching, these versatile "drop-in" replacements offer significant potential for various applications across computer graphics.
Games Summit
Gaming & Interactive
Games
Full Conference Supporter
Full Conference
Experience
DescriptionMotion sickness remains a significant barrier to immersive games, affecting up to one-third of players. We present a player-centered mitigation approach developed within a kinetic first-person game, reframing motion comfort as accessibility and embedding inclusive design directly into gameplay systems, supported by empirical adoption and impact data.
Computer Animation Festival
Full Conference Supporter
Full Conference
DescriptionIn a viking world, a girl’s mother goes to battle following her husband’s death.
But the emotions stemming from this distance haunt them and mysterious wolves appear..
But the emotions stemming from this distance haunt them and mysterious wolves appear..
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Generative AI
Modeling
Full Conference Supporter
Full Conference
DescriptionBFS is a novel generation-based framework for layered image synthesis that produces foreground layers with realistic visual effects while harmonizing with a given background. Using a dual-branch diffusion model and two-stage training with unlayered data, it overcomes data scarcity and achieves superior quality and diversity compared to prior methods.
Talks
New Technologies
Production & Animation
Animation
Dynamics
Geometry
Industry Insight
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionFor Disney/Pixar’s Toy Story 5 character Blaze, we developed a grooming structure for tightly coiled curls, enabling rapid artistic iteration. We deployed Fizt Strands, built on our Fizt cloth solver, offering superior hair collision and stability. These advancements enabled believable curl simulation, complimenting the character’s ethnicity and personality.
Computer Animation Festival
DescriptionJust as he’s about to be placed in the Classics section, a little book named Blooky is accidentally misplaced and ends up lost in the endless shelves of a grand library. From Business and Arts to Politics and even Horror, he explores every aisle, only to be rejected again and again. But when he finally finds a shelf that seems just right, he must decide whether to take his place or help a lonely child who, like him, was cast aside.
Talks
Production & Animation
Research & Education
Animation
Art
Capture/Scanning
Digital Twins
Dynamics
Industry Insight
Lighting
Pipeline Tools and Work
Rendering
Simulation
Full Conference Supporter
Full Conference
DescriptionAvatar: Fire and Ash required geometric consistency between photographed and simulated water in chaotic tank environments over long frame sequences, with characters interacting directly with the surface. We describe a feature-driven approach for recovering temporally coherent water motion to support stable parametric modeling and simulation coupling.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Generative AI
Lighting
Full Conference Supporter
Full Conference
DescriptionBodyReLux is a video diffusion framework for relighting full-body human performances with photorealistic, temporally consistent results. Trained on video relighting pairs captured on large LED Sphere using digital bi-pack technology, it enables precise, dynamic lighting control and generalizes robustly to in-the-wild videos.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionNeural eigenanalysis enables reduced-order modeling across shape families but is limited to Neumann boundaries. We extend it to Dirichlet, Robin, and mixed conditions by treating boundary configurations as inputs, forming a unified shape–boundary space for consistent spectral analysis and simulation across varying geometries and boundary settings.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Lighting
Modeling
Real-Time
Rendering
Full Conference Supporter
Full Conference
DescriptionIn this paper, we present an optimized algorithm for rendering Gaussian Process Implicit Surfaces (GPISes). By using stratified Bernoulli impulses, we precompute point-level bounds stored in a bound map. We also represent region-level bounds with a spatial data structure. Combining these bounds, we achieve significant improvements in rendering performance.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Augmented Reality
Generative AI
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionBoxCtrl introduces a 3D-aware visual prompting framework for precise image editing. Using RGB 3D bounding boxes to decouple geometry from appearance enables accurate translation, rotation, and scaling. A two-stage training of first SFT and then Reinforcement Learning achieves superior precision and photorealism in geometric editing across diverse real-world data.
Talks
Arts & Design
Production & Animation
Animation
Art
Industry Insight
Lighting
Pipeline Tools and Work
Rendering
Full Conference Supporter
Full Conference
DescriptionThe team at Sony Pictures Imageworks developed a scalable compositing workflow for GOAT to achieve a hand-painted aesthetic. Deconstructing illustrations into a "recipe" of plug-and-play modules helped balanced stylization with efficiency. New pipelines for asset-based FX curves and 2D-to-3D sketch translation addressed the show's unique look-of-picture needs.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Full Conference Supporter
Full Conference
DescriptionBrepForge is a generative framework for CAD B-rep modeling that factorizes synthesis into wireframe construction and boundary-conditioned surface generation. This two-stage design improves topological validity and geometric precision, enabling the creation of complex and realistic CAD models.
Educator's Forum
Arts & Design
Gaming & Interactive
Research & Education
Animation
Art
Education
Games
Scientific Visualization
Full Conference Supporter
Full Conference
Experience
DescriptionThis paper reports an eight-year Japan–U.S. collaborative game-making program between an art university and a game program. We present a transferable educational framework that bridges artistic expression and system-oriented game design. Qualitative observation identifies three strategies: role-fluid teams, intent-driven interaction design goals, and the pedagogical use of cross-cultural differences.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Generative AI
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionThis paper introduces a framework for generating high-quality, semantically controllable facial assets. By leveraging a diffusion model's diversity, we develop a face asset synthesis pipeline to produce paired albedo and geometry data with semantic labels. These pairs are then used to train a disentangled GAN, enabling precise latent editing of the generated assets while preserving identity.
Production Session
Production & Animation
Animation
Lighting
Modeling
Performance
Real-Time
Rendering
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionIndustrial Light & Magic invites you behind the scenes of The Mandalorian and Grogu for an in-depth exploration of the visual effects that brought the latest Star Wars feature to life. In this panel, the filmmakers and artists will unpack how groundbreaking real-time technology, virtual production workflows, and time-tested VFX techniques came together to create immersive worlds, unforgettable characters, and epic action on the big screen. From on-set innovation to final pixels, join the ILM team as they share creative insights, technical breakthroughs, and lessons learned while pushing the boundaries of modern visual storytelling.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Full Conference Supporter
Full Conference
DescriptionWe present a broadband hyperspectral 3D (BH3D) imaging system spanning visible to SWIR (450–1500 nm). Using a single-spectrograph stereo setup, our method jointly reconstructs accurate 3D geometry and material properties, enabling robust analysis of real-world scenes beyond the visible spectrum.
Art Paper
Arts & Design
Art
Artificial Intelligence/Machine Learning
Audio
Augmented Reality
Digital Twins
Games
Geometry
Hardware
Performance
Real-Time
Scientific Visualization
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionBruSHŪ is a sensor-embedded calligraphy brush that translates fleeting micro-actions, contact boundaries, turns, and hesitations into sparse sound events and an evolving ink-like visual field. Rather than scoring or reconstructing strokes, it makes tacit embodied decisions perceptible and discussable in real-time, advancing non-normative, cross-modal feedback for reflective creative practice.
Talks
Arts & Design
Production & Animation
Animation
Art
Industry Insight
Lighting
Pipeline Tools and Work
Rendering
Full Conference Supporter
Full Conference
DescriptionFor Pixar’s "Hoppers", the artistic vision called for a distinctive painterly aesthetic that would give the film a handcrafted, illustrative quality while maintaining our sophisticated lighting and color work. This talk presents the Paintbrush Pipeline, a flexible asset and compositing approach developed to balance artistic control with technical requirements.
ACM SIGGRAPH 365
Research & Education
Diversity, Equity and Inclusion
Education
Ethics and Society
Full Conference Supporter
Full Conference
Experience
DescriptionThe ACM SIGGRAPH Rainbow Affinity Group was formally established in June 2025 to support LGBTQ+ individuals in computer graphics and interactive techniques. Participants throughout the years of the Rainbow Meetup Birds of a Feather articulated strong interest in mentorship, networking, and inclusive programming. This session continues that discussion and provides a framework for action.
This session presents the Rainbow Affinity Group as an active organizational design and transition effort. We reflect on early community input, examine the gap between ideation and execution, and introduce a structured model for forming a working committee to support future programming and leadership development.
Through a facilitated discussion, attendees will be invited to help define priorities and identify pathways for participation. The session will initiate the formation of a new contributor cohort aligned with defined roles in programming, mentorship, communications, and industry engagement. This cohort will form the foundation for the group’s next phase and future leadership, in alignment with ACM SIGGRAPH governance practices.
This session presents the Rainbow Affinity Group as an active organizational design and transition effort. We reflect on early community input, examine the gap between ideation and execution, and introduce a structured model for forming a working committee to support future programming and leadership development.
Through a facilitated discussion, attendees will be invited to help define priorities and identify pathways for participation. The session will initiate the formation of a new contributor cohort aligned with defined roles in programming, mentorship, communications, and industry engagement. This cohort will form the foundation for the group’s next phase and future leadership, in alignment with ACM SIGGRAPH governance practices.
Course
Gaming & Interactive
New Technologies
Augmented Reality
Education
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionCreate at the forefront of spatial computing with Meta’s Wearable Device Access Toolkit. This immersive, 3-hour course offers hands-on access to our AI Glasses hardware and sensor data, empowering attendees to build privacy-aware, innovative AI Glasses applications. Participants will advance wearable technology setting new standards for responsible, visionary digital experiences.
Technical Paper
Research & Education
Animation
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionThe Material Point Method struggles with separating materials due to a shared background grid that couples velocities. We introduce a hybrid approach using separate velocity grids and a unified pressure model, enabling buoyancy-driven phase separation for immiscible fluids.
Technical Paper
Research & Education
Geometry
Modeling
Real-Time
Rendering
Full Conference Supporter
Full Conference
DescriptionIn this paper, we generalize the Coons volume from hexahedral topology to arbitrary polyhedral topologies via generalized barycentric coordinates. We prove that the proposed generalized Coons volume possesses several desirable geometric properties and demonstrate its applications in computer graphics.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionCAdam resolves the mismatch between reconstruction-native densification and stochastic generative guidance in optimization-based 3DGS. By reinterpreting densification as statistical signal verification via momentum and context-adaptive selection, it reduces Gaussian primitives by 85%–97% across diverse backbones while preserving comparable perceptual quality, making generative 3DGS far more compact and practical.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Rendering
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionCAGS is a novel color-adaptive system for 3D Gaussian Splatting that enables high-quality, immersive volumetric video streaming. By utilizing vector quantization and reference images for color restoration, CAGS overcomes bandwidth limitations, delivering significantly faster speeds and superior visual quality compared to existing solutions under fluctuating networks.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Generative AI
Full Conference Supporter
Full Conference
DescriptionCameraSquad is a multi-trajectory camera control framework that supports both single-trajectory and parallel multi-trajectory video generation. By decoupling content and camera control mechanisms and using a dual-mode cross-view attention mechanism, it ensures viewpoint consistency and camera control precision in multi-trajectory mode.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Generative AI
Modeling
Full Conference Supporter
Full Conference
DescriptionCanvas-to-Image is a unified framework enhancing diffusion models' compositional and multimodal control through a single canvas interface. By encoding subject references, bounding boxes, and poses into one composite canvas, it enables integrated visual-spatial reasoning. This approach streamlines heterogeneous control signals, allowing models to generate high-fidelity images with precise structural alignment.
Computer Animation Festival
Full Conference Supporter
Full Conference
DescriptionThis film denounces intensive livestock farming by evoking the imprisonment, sadness and resilience of a cow separated from her calf, in a heavy and poignant atmosphere.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Generative AI
Geometry
Modeling
Rendering
Simulation
Full Conference Supporter
Full Conference
DescriptionCaRaFe is the first method to jointly reconstruct 3D scenes from both camera and radar data using a multi-modal shared neural geometry representation. It can reconstruct large-scale urban environments from a single drive-through and synthesize measurements from both sensors at novel viewpoints, leveraging their complementary strengths to improve geometric fidelity.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Generative AI
Geometry
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionCasLayout is a cascaded diffusion framework that decomposes 3D indoor scene generation into four interpretable stages. This design reduces data requirements while enabling flexible, controllable synthesis and seamless integration with LLMs and VLMs for zero-shot tasks.
ACM SIGGRAPH 365
Arts & Design
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Animation
Art
Artificial Intelligence/Machine Learning
Audio
Augmented Reality
Capture/Scanning
Computer Vision
Digital Twins
Display
Dynamics
Education
Ethics and Society
Fabrication
Games
Generative AI
Geometry
Graphics Systems Architecture
Haptics
Hardware
Image Processing
Industry Insight
Lighting
Math Foundations and Theory
Modeling
Performance
Physical AI
Pipeline Tools and Work
Real-Time
Rendering
Robotics
Scientific Visualization
Simulation
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionCelebrate the start of SIGGRAPH with an evening you won't want to miss! Join the ACM SIGGRAPH Chapters Committee from 9 pm to 1 am on Monday, July 20 at Exchange LA for a night of fun with the world’s greatest pixel wranglers. All registration levels are welcome, and your conference badge is your entry ticket! You must be of legal drinking age to attend and provide a valid photo ID in order to enter the club. The venue is a 10 minute car ride away from the Los Angeles Convention Center at 618 S Spring Street, Los Angeles, CA. Ask for more information at the ACM SIGGRAPH Village!
ACM SIGGRAPH 365
Arts & Design
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Animation
Art
Artificial Intelligence/Machine Learning
Audio
Augmented Reality
Capture/Scanning
Computer Vision
Digital Twins
Display
Dynamics
Education
Ethics and Society
Fabrication
Games
Generative AI
Geometry
Graphics Systems Architecture
Haptics
Hardware
Image Processing
Industry Insight
Lighting
Math Foundations and Theory
Modeling
Performance
Physical AI
Real-Time
Rendering
Robotics
Scientific Visualization
Simulation
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionIn conjunction with the Professional and Student Chapters Committee (PSCC), leaders from chapters on the East Coast of North America welcome you to join a social hour to connect with other students and professionals. This session will give information on chapters in the area, the activities they host, and ways to get involved.
More information about the PSCC, chapters near you and how to get involved can be found at https://siggraph.org/chapters.
More information about the PSCC, chapters near you and how to get involved can be found at https://siggraph.org/chapters.
ACM SIGGRAPH 365
Arts & Design
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Animation
Art
Artificial Intelligence/Machine Learning
Audio
Augmented Reality
Capture/Scanning
Computer Vision
Digital Twins
Display
Dynamics
Education
Ethics and Society
Fabrication
Games
Generative AI
Geometry
Graphics Systems Architecture
Haptics
Hardware
Image Processing
Industry Insight
Lighting
Math Foundations and Theory
Modeling
Performance
Physical AI
Pipeline Tools and Work
Real-Time
Rendering
Robotics
Scientific Visualization
Simulation
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionIn conjunction with the Professional and Student Chapters Committee (PSCC), leaders from chapters on the West Coast of North America welcome you to join a social hour to connect with other students and professionals. This session will give information on chapters in the area, the activities they host, and ways to get involved.
More information about the PSCC, chapters near you and how to get involved can be found at https://siggraph.org/chapters.
More information about the PSCC, chapters near you and how to get involved can be found at https://siggraph.org/chapters.
Production Session
Production & Animation
Animation
Art
Geometry
Image Processing
Industry Insight
Lighting
Modeling
Pipeline Tools and Work
Rendering
Simulation
Full Conference Supporter
Full Conference
DescriptionWhen building upon a beloved franchise like Zootopia, the artists of Walt Disney Animation Studios had to consider every detail that makes the world unique and its characters memorable. By combining the famous bunny and fox duo with a herd of new animated animals and environments, artists developed new rules and techniques to expand the world in genuine and exciting ways.
In this session, the “Zootopia 2” leadership team will take you on a wild journey through the film’s worldbuilding. They will explore what it took to visualize marshes, mountains, deserts, and tundra, and discuss the development of new characters like Gary De’Snake and the Lynxley family. They will also share new workflows and tools throughout the discussion that allowed for collaboration across multiple departments, and enabled the crew to tackle an adventurous and expansive body of work.
In this session, the “Zootopia 2” leadership team will take you on a wild journey through the film’s worldbuilding. They will explore what it took to visualize marshes, mountains, deserts, and tundra, and discuss the development of new characters like Gary De’Snake and the Lynxley family. They will also share new workflows and tools throughout the discussion that allowed for collaboration across multiple departments, and enabled the crew to tackle an adventurous and expansive body of work.
Spatial Storytelling
Gaming & Interactive
Art
Games
Real-Time
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionThis talk explores "cheap and cheerful" prototyping in immersive storytelling through the evolution of THE FOOTAGE, a VR narrative horror game. It traces how low-fidelity, no-code, and provisional methods surfaced crucial spatial insights, guided design decisions, enabled play-testing, and helped secure funding before full-scale production.
Talks
Arts & Design
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Art
Artificial Intelligence/Machine Learning
Audio
Computer Vision
Digital Twins
Education
Fabrication
Games
Generative AI
Geometry
Industry Insight
Modeling
Performance
Pipeline Tools and Work
Real-Time
Rendering
Full Conference Supporter
Full Conference
DescriptionChem2Vid introduces a multimodal framework translating chemical equations into dynamic, chemically faithful videos for education. It integrates LLMs and chemical simulators with diffusion models to generate stoichiometry-constrained reaction visualizations and synchronized property curves, creating a low-cost virtual laboratory to enhance chemistry understanding.
Technical Paper
Research & Education
Fabrication
Geometry
Modeling
Robotics
Simulation
Full Conference Supporter
Full Conference
DescriptionA field-based computational framework for designing fabric-reinforced composites with optimized topology and manufacturable semi-continuous layers, producing laminates up to 43.8% stiffer than conventional planar-ply designs.
Educator's Forum
Arts & Design
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Animation
Art
Augmented Reality
Digital Twins
Education
Games
Generative AI
Real-Time
Simulation
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionBridging visual and syntax-based programming, this Art-Thinking curriculum fosters intrinsic motivation. By combining Apple’s SF Symbols to procedurally generate original Japanese patterns (Wagara), students master abstract concepts like coordinates and loops. This culturally responsive approach transforms syntax anxiety into "productive struggle," yielding tangible, creative outcomes.
Technical Paper
Research & Education
Augmented Reality
Display
Fabrication
Hardware
Real-Time
Rendering
Simulation
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionWe propose CoherentRaster, an efficient 3D Gaussian Splatting-based light field rendering framework for glasses-free 3D displays. CoherentRaster introduces Cross-view Coherent Attribute Reuse to eliminate redundant computations across neighboring viewpoints and View-coherent Remapping to restore warp-level memory efficiency degraded by interlaced subpixel layouts, enabling real-time, high-quality synthesis.
Educator's Forum
Gaming & Interactive
New Technologies
Research & Education
Augmented Reality
Digital Twins
Education
Games
Image Processing
Real-Time
Scientific Visualization
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionCollabXR: a collaborative, multi-user XR platform that enables instruction, training and research through visualization and interaction with immersive three-dimensional materials. Originally designed for sharing astrophysics visualizations of supernova remnants in an 'XR classroom,' the platform has expanded to incorporate other 3D data from STEM, arts, humanities and cultural heritage.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Generative AI
Modeling
Full Conference Supporter
Full Conference
DescriptionColorful-Noise is a training-free image generation conditioning method that leverages FFT noise frequency manipulation. We show that low frequencies dominate output appearance and can encode structure from scribbles, color maps, or reference images — enabling anyone to condition image generation with zero overhead, no training, and no significant artistic skill.
Talks
Arts & Design
Production & Animation
Animation
Art
Industry Insight
Lighting
Pipeline Tools and Work
Rendering
Full Conference Supporter
Full Conference
DescriptionCreating Toy Story 5's hand-drawn "Playtime" sequences ventured outside Pixar's traditional pipeline via collaborative non-linear workflows spanning effects, shading, and compositing. The team developed pre-lit geometric pastel assets, stylized character treatments, and layered compositing approaches that transformed CG animation into ethereal pastel worlds visualizing childhood imagination.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Generative AI
Modeling
Full Conference Supporter
Full Conference
DescriptionWe study an important factor in diffusion generative models: combinatorial complexity. Existing training schemes insufficiently sample the joint space of dimensions and attributes, leading to weaker test time performance. We address this with ComboStoc, a simple stochastic process that better exploits underlying combinatorial structure.
Technical Paper
Research & Education
Augmented Reality
Display
Fabrication
Hardware
Real-Time
Rendering
Simulation
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionComplex-valued holographic radiance fields optimize holographic 3D scenes without relying on intensity-based intermediaries. Leveraging multi-view images, our method uses complex-valued Gaussian primitives encoding amplitude and phase aligned with scene geometry. This eliminates expensive single-view hologram recalculation, accelerating speed by 30×–10,000× while matching state-of-the-art quality, bridging wave optics and 3D scene geometry.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionTo solve ultra-large-scale modal analysis problems, we extend Component Mode Synthesis (CMS) methods to use multiple partitions to generate more efficient local eigenbasis functions and reduce the computational cost of the low-frequency interface modes. We obtained the lowest 1500 eigenmodes of a model (30M DoFs) in 88 minutes.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Generative AI
Modeling
Full Conference Supporter
Full Conference
DescriptionWe present a method for generating realistic multi-person interaction scenes from text. By modeling human pose and appearance together and composing scenes step by step, our approach produces images that better capture complex interactions, with improved accuracy and greater diversity compared to existing text-to-image models.
Technical Paper
Research & Education
Fabrication
Geometry
Modeling
Robotics
Simulation
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Technical Paper
Research & Education
Fabrication
Geometry
Modeling
Robotics
Simulation
Full Conference Supporter
Full Conference
DescriptionWe present a computational approach for designing contact-based coordinate-motion assemblies that meet user-specified target appearance and motion. One key enabler of our approach is that we established a theoretical connection between contact geometry within an assembly and unique coordinate motion of the assembly's parts.
Technical Paper
Research & Education
Fabrication
Geometry
Modeling
Robotics
Simulation
Full Conference Supporter
Full Conference
DescriptionThe interplay of morphology, control, and anisotropic friction makes optimal design for terrestrial locomotion challenging. We present a computational pipeline that co-designs friction and controllers across diverse robot morphologies, showing anisotropic friction is critical and achieving statistically superior performance over state-of-the-art methods.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Art Paper
Arts & Design
Art
Computer Vision
Education
Fabrication
Generative AI
Geometry
Hardware
Image Processing
Math Foundations and Theory
Performance
Real-Time
Robotics
Full Conference Supporter
Full Conference
Experience
DescriptionThis study investigates Frider Nake's 1970 programming language "L" to explore the intersection of code and art. By examining how "L" translates programming languages into imaginative and intellectual explorations, we gain insights into the cultural dimensions of code and its potential as a medium for artistic production. This analysis deepens our understanding of how programming languages historically engaged with images, graphics and visual semiotics.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Fabrication
Geometry
Modeling
Full Conference Supporter
Full Conference
DescriptionWe present a simple discretization of constant mean curvature surfaces based on the classical observation that their Gauss maps are harmonic. Our construction is elementary---requiring only discrete Dirichlet energy minimization and a Poisson solve---yet it exactly mirrors this aspect of the smooth theory.
Technical Paper
Research & Education
Fabrication
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionWe present two novel methods for automatically enhancing the continuity of piecewise-polynomial curves with negligible computational overhead. They provide the first practical interface for modeling curves with curvature (or higher) continuity using the familiar Bezier handles of piecewise cubic curves, the most popular curve modeling interface that previously offered no mechanism to achieve curvature continuity. Our solutions can also be used for defining new interpolating curve formulations with desirable properties. Furthermore, they can be used with higher-dimensional curves or surfaces, which we demonstrate by fixing the curvature discontinuities of the famous Utah Teapot model.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Augmented Reality
Generative AI
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionDiffusion editing models lack smooth control over edit intensity, and scaling standard Classifier-Free Guidance fails to provide it. We introduce Adaptive-Origin Guidance (AdaOr), which dynamically adjusts the guidance origin using an identity-conditioned prediction. This enables continuous, fine-grained control over both image and video models at inference time without specialized datasets.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Modeling
Real-Time
Rendering
Full Conference Supporter
Full Conference
DescriptionWe introduce a novel Diffusion Transformer framework for high-fidelity, controllable texture tiling. By leveraging Coordinate-Transformed Rotary Embeddings, our method enables precise manipulation of texture frequency, scale, and orientation. It seamlessly integrates reference patterns into images while perfectly preserving the original scene's lighting and geometry.
Immersive Pavilion
Arts & Design
Gaming & Interactive
New Technologies
Research & Education
Animation
Art
Artificial Intelligence/Machine Learning
Audio
Augmented Reality
Computer Vision
Digital Twins
Education
Ethics and Society
Fabrication
Games
Generative AI
Haptics
Hardware
Image Processing
Modeling
Performance
Pipeline Tools and Work
Real-Time
Scientific Visualization
Simulation
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionFrom the award-winning storytellers at Atlantic Studios, Cosmos Unseen: Black Holes is a first-of-its-kind immersive experience that demystifies complex spacetime phenomena. Crafted for the launch of the Google x Samsung AndroidXR headset, Cosmos Unseen is a breakthrough in scientific storytelling, letting users get hands-on with the mysteries of our universe.
Talks
New Technologies
Production & Animation
Animation
Dynamics
Geometry
Industry Insight
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionWe present the feather system for Disney/Pixar's "Hoppers," covering multiple bird species. We detail customizing Houdini's hair grooming pipeline, integrating groomed feathers into Presto character rigs, developing a flexible shading framework supporting patterns and two-sided coloration, and creating procedural and simulated flutter effects for natural animation and wind response.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Talks
Production & Animation
Research & Education
Animation
Art
Capture/Scanning
Digital Twins
Dynamics
Industry Insight
Lighting
Pipeline Tools and Work
Rendering
Simulation
Full Conference Supporter
Full Conference
DescriptionThe Flux Devil, an elemental vortex in the climactic sequence of Avatar: Fire and Ash, is made of plasma, water and fire. To create an art-directable effect, while adhering to the laws of nature, we implemented specialized emission tools, warped gravity fields for buoyant updrafts and spatially varying Coriolis forces.
Talks
New Technologies
Production & Animation
Animation
Dynamics
Geometry
Industry Insight
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionCapturing the look and performance of Gary De’Snake’s scales in an appealing way was a key challenge in Disney’s “Zootopia 2”. New procedural workflows for creating and deforming scales made it possible for artists to efficiently achieve the desired look/motion while allowing for specific art-direction and exaggerated animation performance.
Talks
Arts & Design
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Animation
Art
Artificial Intelligence/Machine Learning
Ethics and Society
Generative AI
Industry Insight
Pipeline Tools and Work
Full Conference Supporter
Full Conference
DescriptionWe introduce SIGGRAPH to critical theories of generative models, focusing on socioeconomic relations to research and art. We introduce critical theory through computing's relationship to commodification and labor and contrast two views of generative models: one as the automation of labor and another examining their role in altering human sociality.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Full Conference Supporter
Full Conference
DescriptionCubePart is an open-vocabulary 3D generative framework offering explicit semantic part-based control over mesh structure. Guided by a text prompt and custom part schema, it synthesizes coherent multi-part meshes, which are ready for direct integration into game engines and animation pipelines without manual post-processing.
Educator's Forum
Arts & Design
Production & Animation
Research & Education
Animation
Education
Industry Insight
Pipeline Tools and Work
Full Conference Supporter
Full Conference
Experience
DescriptionThis panel brings together recruiting leaders from major animation and VFX studios to share first-person insight into contemporary hiring realities. Through candid discussion, panelists clarify evolving skill demands, internship and entry-level pathways, and recruitment processes, offering artists, students, and educators a clearer, more realistic snapshot of an industry in transition.
Computer Animation Festival
Full Conference Supporter
Full Conference
DescriptionHedera, a mother fairy, is foretold that her daughter Pavera would be killed by a dark being. Hedera isolates her child on an island, overprotecting her, while the young girl grows more and more frustrated. After an argument between them, Pavera sees a star falling from the sky at night and runs off. She encounters Ophrys, a fairy wearing dark clothes. Hedera wakes up, finds Pavera with Ophrys and attacks the latter on sight. Pavera intervenes and both young people are killed by Hedera's last blow. Covered in ashes, Hedera realizes she is the being of the prophecy and gives up on living.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionWe use the material curvatures of the super-helix model to better analyze and edit tightly coiled hair. Within this curvature space, we propose novel operations for hair ruffling, heuristics for tracking flips in handedness, and control armatures that enable intuitive ensemble strand editing.
Course
Production & Animation
Research & Education
Animation
Math Foundations and Theory
Full Conference Supporter
Full Conference
DescriptionThis course is an introduction to curves for technical artists and aspiring graphics engineers. We will build an understanding of curves used in computer graphics, especially rigging, from simple mathematical notation, using graphical visualisations, code implementation and relating them to their use in DCCs.
ACM SIGGRAPH 365
Arts & Design
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Animation
Art
Augmented Reality
Education
Games
Performance
Real-Time
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionThis session offers a preview of SIGGRAPH 2026’s creative programming. Chairs and Directors from across the conference will share highlights from their programs, including featured projects, upcoming events, and curatorial directions. Attendees will gain an early view of how this year’s conference is expanding the role of digital art, encouraging cross-disciplinary exchange, and supporting experimentation across media, technology, and artistic practice.
The session also introduces the ACM SIGGRAPH Digital Arts Community (DAC), a platform for global dialogue among artists, designers, and technologists. Through initiatives such as SPARKS Lightning Talks, curated digital exhibitions, and student competitions, DAC connects artistic practice with computer graphics, interactive media, and emerging technologies. More information is available at dac.siggraph.org.
The session also introduces the ACM SIGGRAPH Digital Arts Community (DAC), a platform for global dialogue among artists, designers, and technologists. Through initiatives such as SPARKS Lightning Talks, curated digital exhibitions, and student competitions, DAC connects artistic practice with computer graphics, interactive media, and emerging technologies. More information is available at dac.siggraph.org.
ACM SIGGRAPH 365
Arts & Design
Production & Animation
Animation
Art
Artificial Intelligence/Machine Learning
Generative AI
Performance
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionThis session will present the highlights from the fall 2025 and spring 2026 SPARKS (Short Presentations for the Kindred Spirit) sessions. Speakers from seven previous SPARKS session will show examples of their work and discuss the research behind their work. The "Entangled (Im)materialities" session moderated by Clea T. Waite and Bonnie Mitchell in Nov. explored artworks that challenged the structures of perception, knowledge, memory, and materiality. "Embodied Convergences: Humans, Machines, and Performance", moderated by Mona Kasra in Dec. featured artists and researchers who are redefining the possibilities of human–machine collaboration through embodied, computational, and performative practices. "Beyond the Global North: Southeast Asia’s New Wave of Generative and Interactive Innovation" moderated by Patrick Hartono and Bin You in Jan. included artists who are shaping new trajectories for generative and interactive visual culture in the SE Asia. "Interabilities in 21st Century Arts & Technology", moderated by James Hullick and Melentie Pandilovski in Feb. included neurodiverse artists who are reshaping contemporary society. "Alter Nature: Exploring Life through Computational Arts", moderated by Zi-Wei Wu in Mar. included pioneers and young artists working with the concept of artificial life. "Confluence of Generative AI and the Art of Filmmaking", moderated by Sal Ardongdé and Paul Trillo in Apr., explored how generative AI is transforming workflows and creative outcomes in the film industry. "Animism Revisited: Worlding with Digital and More-than-Human Minds", moderated by Michael Just and Gustavo Alfonso Rincon in May, considered how creative practices are reopening and complicating animist ways of sensing, valuing, and coexisting.
Art Paper
Arts & Design
Art
Computer Vision
Education
Fabrication
Generative AI
Geometry
Hardware
Image Processing
Math Foundations and Theory
Performance
Real-Time
Robotics
Full Conference Supporter
Full Conference
Experience
DescriptionDefining data materialization is relevant because it is an emerging practice that like SIGGRAPH lies at the intersection of art and technology. Furthermore, as data collection explodes in every facet of society, data materialization offers a new form of epistemic translation--the transformation of abstract datasets into physical forms that generate meaning through material–data connections rather than informational communication.
Computer Animation Festival
Full Conference Supporter
Full Conference
DescriptionAn old postman goes through his last tour of deliveries in the archipelago that he worked in his whole life, gliding on water with his old bike. He finds himself confronted to the reality of his coming retirement.
Computer Animation Festival
Full Conference Supporter
Full Conference
DescriptionA young woman desperate for a good night’s sleep is kept awake by her exceedingly noisy neighbors. As she struggles to imagine what could be causing the cacophony upstairs, reality drifts into fantasy, and an epic battle for peace and sanity ensues.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Robotics
Simulation
Full Conference Supporter
Full Conference
DescriptionWe present a deep motion warping framework utilizing a phase-conditioned diffusion autoencoder. By explicitly disentangling motion into root velocity, phase, and style, it enables intuitive, high-level character animation editing including root warping, exaggeration, time warping, and style transfer while strictly preserving structural consistency.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Fabrication
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionWe introduce DeepMill++, a conservative and highly efficient framework for cutter accessibility analysis on arbitrary triangular meshes. DeepMill++ reformulates accessibility and occlusion detection as a rasterization-based visibility and depth pooling problem. DeepMill++ achieves up to 9.5× speedup over state-of-the-art geometric methods, while maintaining 97.5% conservative accessibility accuracy.
Emerging Technologies
Arts & Design
Gaming & Interactive
New Technologies
Research & Education
Art
Artificial Intelligence/Machine Learning
Audio
Augmented Reality
Capture/Scanning
Computer Vision
Digital Twins
Display
Dynamics
Education
Fabrication
Games
Generative AI
Graphics Systems Architecture
Haptics
Hardware
Image Processing
Performance
Physical AI
Pipeline Tools and Work
Real-Time
Robotics
Scientific Visualization
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionWe present a set of redesigned CNC machines for craft-aligned digital fabrication, that combine pre-planned and real-time motion control.
These machines re-envision how programmatic machine control can support a diversity of fabrication workflows, in particular demonstrating the potential for machine designers to include real-time physical input.
These machines re-envision how programmatic machine control can support a diversity of fabrication workflows, in particular demonstrating the potential for machine designers to include real-time physical input.
Appy Hour
Arts & Design
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Animation
Art
Artificial Intelligence/Machine Learning
Audio
Augmented Reality
Computer Vision
Digital Twins
Education
Ethics and Society
Games
Generative AI
Geometry
Image Processing
Math Foundations and Theory
Modeling
Performance
Physical AI
Pipeline Tools and Work
Real-Time
Robotics
Scientific Visualization
Simulation
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionThis project positions game development as a "translation device" to integrate Quantum Annealing into society. It transforms abstract quantum computing into intuitive interactive experiences, validating its social implementation by bridging complex technology and human everyday perception through engaging gameplay and visuals.
Educator's Forum
Arts & Design
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Animation
Art
Augmented Reality
Capture/Scanning
Education
Ethics and Society
Generative AI
Math Foundations and Theory
Modeling
Performance
Pipeline Tools and Work
Real-Time
Rendering
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionThis talk describes three projects based on motion tracking that help students link their intuition about the real-world to theoretical concepts, such as frames of reference and kinematics. The projects involve hands-on construction and puppetry and also allow students to show their creativity.
Emerging Technologies
Arts & Design
Gaming & Interactive
New Technologies
Research & Education
Art
Artificial Intelligence/Machine Learning
Audio
Augmented Reality
Capture/Scanning
Computer Vision
Digital Twins
Display
Dynamics
Education
Fabrication
Games
Generative AI
Graphics Systems Architecture
Haptics
Hardware
Image Processing
Performance
Physical AI
Pipeline Tools and Work
Real-Time
Robotics
Scientific Visualization
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionWe present XReactor, a compact motor that creates realistic 3D vibrations in any direction. Unlike traditional composite devices, its simple design provides 3d haptic sensations without complex controls. By combining it with VR, users can feel the more realistic weight and resistance of virtual objects.
Technical Paper
Research & Education
Fabrication
Geometry
Modeling
Robotics
Simulation
Full Conference Supporter
Full Conference
DescriptionDiceplay is a modular, physical, low-resolution canvas that uses identical geometric dice to create abstract visual compositions. We propose a gradient-based optimization approach to generate Diceplay configurations from text prompts by using a novel shape grammar and relaxing the purely discrete design space to a mixed-discrete-continuous problem.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Augmented Reality
Rendering
Simulation
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionThis paper presents a psychophysical study of dichoptic foveation, applying blur to one eye and sharpening to the other, in virtual reality. A perceptual model of interocular frequency differences across the visual field is developed and applied to enhance foveated rendering quality in VR headsets.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Modeling
Rendering
Simulation
Full Conference Supporter
Full Conference
DescriptionWe present a differentiable Monte Carlo framework for multi-energy group neutron transport, extending ideas from differentiable rendering to nuclear engineering. The method enables gradient-based sensitivity analysis and inverse design.
Technical Workshop
Arts & Design
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Animation
Artificial Intelligence/Machine Learning
Capture/Scanning
Computer Vision
Digital Twins
Dynamics
Education
Fabrication
Games
Generative AI
Geometry
Industry Insight
Modeling
Physical AI
Robotics
Simulation
Spatial Computing
Full Conference Supporter
Full Conference
DescriptionDifferentiable Physics for Graphics and AI (DPGA) is a half-day SIGGRAPH technical workshop exploring how differentiable physics-based simulation is transforming graphics and related fields including robotics, 3D vision, design, and fabrication. As simulation becomes tightly integrated with optimization, learning, and generative pipelines, differentiability enables new workflows for inverse problems, system identification, and physics-aware design.
The workshop is organized by researchers from Carnegie Mellon University, UCLA, the University of Utah, and Meta Reality Labs, whose work spans physical simulation, graphics systems, robotics, and AI. Together with invited speakers from academia and industry, they bring perspectives from foundational simulation research to large-scale production systems and real-world applications.
The workshop brings together foundational methods, practical systems, and emerging applications in differentiable simulation across graphics, robotics, 3D vision, design, and fabrication. Through keynote talks, lightning presentations, and a moderated panel, it will showcase both what differentiable physics can already enable and the technical barriers that still limit broader adoption, including efficiency, robustness, contact-rich dynamics, and the gap between application-facing representations and simulation-ready assets.
Attendees can expect an engaging forum that connects developers with researchers and practitioners seeking to use simulation as an optimization and learning primitive, not just a forward model. A key merit of the workshop is its ability to bridge communities that do not often meet in one venue, helping participants identify practical opportunities, clarify open problems, and shape future directions for differentiable simulation in graphics and AI.
The workshop is organized by researchers from Carnegie Mellon University, UCLA, the University of Utah, and Meta Reality Labs, whose work spans physical simulation, graphics systems, robotics, and AI. Together with invited speakers from academia and industry, they bring perspectives from foundational simulation research to large-scale production systems and real-world applications.
The workshop brings together foundational methods, practical systems, and emerging applications in differentiable simulation across graphics, robotics, 3D vision, design, and fabrication. Through keynote talks, lightning presentations, and a moderated panel, it will showcase both what differentiable physics can already enable and the technical barriers that still limit broader adoption, including efficiency, robustness, contact-rich dynamics, and the gap between application-facing representations and simulation-ready assets.
Attendees can expect an engaging forum that connects developers with researchers and practitioners seeking to use simulation as an optimization and learning primitive, not just a forward model. A key merit of the workshop is its ability to bridge communities that do not often meet in one venue, helping participants identify practical opportunities, clarify open problems, and shape future directions for differentiable simulation in graphics and AI.
Technical Paper
Research & Education
Fabrication
Geometry
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionWe provide an efficient technique to differentiate voxel values with respect to the positions of the surface that generated the voxel grid. This allows to solve problems, such as intersection detection, that can be easily solved on regular voxel grids for surface representations such as triangle meshes.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Real-Time
Rendering
Scientific Visualization
Simulation
Full Conference Supporter
Full Conference
DescriptionWe present DiffSurFlow, an efficient and robust framework for long-horizon and vortex-rich fluid optimization. Powered by a physics-informed surrogate gradient strategy that exploits flow map "gradient highways" while pruning computationally intensive high-order gradient passes, it achieves significant speedups and memory reductions while maintaining near-identical gradient accuracy.
Art Gallery
Arts & Design
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Full Conference Supporter
Full Conference
Experience
DescriptionDiffusion TV is an interactive interface that attempts to demystify the hidden processes of AI diffusion models through a nostalgic CRT TV. By manipulating the TV’s knobs and antenna, viewers explore AI-generated images and sounds, experiencing the transformation of chaotic data into recognizable forms. Featuring three channels, Past (extinct animals), Present (endangered species), and Future (speculative AI-generated creatures), the work prompts reflection on the interconnected relationships between humanity, technology, and the environment. Through its blend of artistic interpretation and experiential AI, Diffusion TV invites audiences to consider what is preserved and what is lost in the pursuit of technological advancement.
Educator's Forum
Arts & Design
Gaming & Interactive
Research & Education
Animation
Art
Education
Games
Scientific Visualization
Full Conference Supporter
Full Conference
Experience
DescriptionThe Digital Arts B.A. is a computer science and studio art interdisciplinary degree designed to prepare students for fields such as mobile-application and web design, game development, visual effects, interactive media, graphics, simulation, human-computer interaction and other emerging opportunities. We cover lessons learned on supporting student success in this space.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Generative AI
Modeling
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Talks
Production & Animation
Animation
Modeling
Pipeline Tools and Work
Simulation
Full Conference Supporter
Full Conference
DescriptionIn Disney’s “Zootopia 2”, thousands of mammals gather at a Desert Festival. The expansive
layout of the set, alongside the need for art-directed performances, required an efficient,
multi-scale crowds approach. Environment-driven systems for large-scale movement,
automated placement of authored group performances, and procedural character animation, all
proved effective at scale.
layout of the set, alongside the need for art-directed performances, required an efficient,
multi-scale crowds approach. Environment-driven systems for large-scale movement,
automated placement of authored group performances, and procedural character animation, all
proved effective at scale.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionSDFRaster combines the efficiency of rasterization with signed distance fields for end-to-end mesh reconstruction from multi-view images. By optimizing a tetrahedral SDF and extracting meshes during optimization, it produces accurate, complete, and compact surfaces without heuristic post-processing.
Technical Paper
Research & Education
Animation
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionAffine Body Dynamics within Incremental Potential Contact enables accurate simulation of near-rigid, extremely stiff solids with strict non-penetration. We present a distributed ABD solver based on consensus ADMM, where nodes solve local subproblems in parallel and synchronize shared bodies through global consensus, achieving robust convergence, non-penetration, and scalable multi-node performance.
Technical Paper
Research & Education
Animation
Geometry
Hardware
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionDivide and Truncate (DAT) is a unified framework for penetration-free coupling of multi-physics systems, including rigid bodies, volumetric soft bodies, thin shells, rods, and animated objects. DAT eliminates artificial damping and deadlock issues. Material- and solver-agnostic, DAT plugs into any iterative optimizer as a post-processing step.
Technical Paper
Research & Education
Geometry
Modeling
Real-Time
Rendering
Full Conference Supporter
Full Conference
DescriptionA new method for computing base meshes for displacement mapping, exploiting the Jacobian of the displacement function to guide base-mesh computation. DJM outperforms prior art in terms of reconstruction accuracy-to-size trade off and can be used for both traditional micromesh-based rendering and neural encoding.
Spatial Storytelling
Arts & Design
New Technologies
Art
Artificial Intelligence/Machine Learning
Generative AI
Performance
Physical AI
Real-Time
Robotics
Spatial Computing
Full Conference Supporter
Full Conference
Experience
DescriptionBFF: Dog Walk explores human-AI intimacy through co-parenting two robot dogs. With video and live performance, the artist-researchers challenge the mind-body divide, documenting the evolution of their dogs' LLMs and their life in the intimate everyday. Blending human and machine perspectives, this project seeks a hybrid language exploring the boundaries of authenticity, embodiment, and synthetic companionship.
Technical Paper
Research & Education
Games
Real-Time
Rendering
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionWe introduce a novel Polyphase Filtering framework offering the fastest approximation for real-time, large-kernel image and video processing. Overcoming fixed-resolution processing constraints, our method efficiently handles complex, spatially-variant kernels. It significantly outperforms existing techniques, delivering unmatched computational speed and superior visual quality for dynamic post-processing and advanced visual effects.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Generative AI
Modeling
Full Conference Supporter
Full Conference
DescriptionWe present DreamActor-M2, a universal character animation framework that regards motion conditioning as spatiotemporal in-context learning. It leverages video foundation model priors and enables end-to-end pose-free motion transfer from raw videos. Without explicit pose estimation, it achieves strong generalization and high-fidelity animation across complex scenes.
Art Gallery
Arts & Design
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Full Conference Supporter
Full Conference
Experience
DescriptionDreaming Babies: illi presents 23 speculative human babies, each linked to a different chromosomal variation, sleeping and dreaming in virtual incubators in space. Using AI-generated narratives and interactive sound input, the installation invites audiences to engage with speculative futures of human birth shaped by genetic technology. Inspired by the artist’s experience with PGT, the work questions notions of “normality,” perfection, and genetic selection. Reimagining chromosomal differences not as defects but as poetic identities and superpowers, the project reframes genetic variation as a source of imagination, multiplicity, and beauty.
Talks
Production & Animation
Animation
Modeling
Pipeline Tools and Work
Simulation
Full Conference Supporter
Full Conference
DescriptionTo meet the demands of massive, diverse worlds, Wardrobe decouples character design from technical rigging. By utilizing automated pipelines, it reduces manual setup from days into minutes. This system enables high-quality, scalable character variation, allowing artists to populate entire worlds while reducing technical overhead.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Full Conference Supporter
Full Conference
DescriptionWe propose an algorithm to reconstruct explicit polygonal meshes from discretely sampled Signed Distance Function (SDF) data, which is especially effective at recovering sharp features. Building on the Dual Contouring of Hermite Data method, we solve a quadratic optimization problem to place the mesh's vertices. Critically, this optimization relies solely on discretely sampled SDF data, without requiring arbitrary access to the function, gradient information, or training on large-scale datasets. Our method sets a new state of the art in surface reconstruction from SDFs at medium and high resolutions, and opens the door for applications in 3D modeling and design.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Full Conference Supporter
Full Conference
DescriptionDCx extends the Dual Contouring method which only extracts manifold surfaces from signed distance fields, to enable the surface extraction from unsigned distance fields. This is achieved by looking up a carefully designed table over expanded 2x2x2 cubes, allowing complex topologies to be extracted, particularly for non-manifold structures.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Full Conference Supporter
Full Conference
DescriptionDualBrep presents a novel continuous representation for CAD models, resolving modeling challenges caused by heterogeneous data structures. It uses dual scalar fields: a Signed Distance Function for geometry and an Unsigned Distance Field for topology. Compressing these into a single latent space enables robust reverse engineering and generative modeling.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Rendering
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Augmented Reality
Rendering
Simulation
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionThis work investigates visual dominance as a dynamic property across the FOV. Using a blur detection task, we confirm dynamic dominance in the fovea and calibrate monocular dominance in the periphery. The resulting gaze-independent peripheral dominance map can be integrated into different rendering pipelines to enhance foveation in non-dominant regions.
Technical Paper
Research & Education
Lighting
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionEAG-PT reconstructs indoor scenes with 2D Gaussians for editable diffuse global illumination. By separating emissive and non-emissive components and combining efficient reconstruction with path tracing, it enables more natural, physically consistent scene editing than radiance-field methods, while avoiding the geometry limitations of mesh-based inverse rendering.
ACM SIGGRAPH 365
Early Career Development: Kickstart a Career in Computer Graphics – Resume, Reel & Industry Insights
2:15pm - 3:15pm PDT Sunday, 19 July 2026 Room 404 ABResearch & Education
Education
Industry Insight
Pipeline Tools and Work
Full Conference Supporter
Full Conference
Experience
DescriptionThe ACM SIGGRAPH Early Career Development Committee (ECDC) Resume and Reel Review program has long supported students and emerging professionals through mentorship, portfolio feedback, and year-round career development programming.
In this interactive panel discussion, ECDC brings together industry professionals from fashion, enterprise visualization, infrastructure, product development, and creative technology to discuss the many pathways available within modern 3D careers beyond traditional entertainment pipelines.
Panelists include leaders working in apparel visualization, immersive media, enterprise 3D workflows, AI-assisted production, and digital product creation. Together, they will share practical insights into resumes and demo reels, hiring expectations, portfolio strategy, networking, mentorship, and adapting creative skills across evolving industries and technologies.
Rather than a traditional lecture, this session is designed as an interactive conversation shaped by audience questions and discussion. Attendees will have the opportunity to engage directly with panelists on topics including career transitions, breaking into new industries, building sustainable creative careers, and navigating the rapidly evolving landscape of 3D tools and workflows.
Whether attendees are students, emerging professionals, or experienced artists exploring new directions, this session offers an opportunity to gain real-world insight from artists and technologists actively shaping the future of 3D across multiple industries.
In this interactive panel discussion, ECDC brings together industry professionals from fashion, enterprise visualization, infrastructure, product development, and creative technology to discuss the many pathways available within modern 3D careers beyond traditional entertainment pipelines.
Panelists include leaders working in apparel visualization, immersive media, enterprise 3D workflows, AI-assisted production, and digital product creation. Together, they will share practical insights into resumes and demo reels, hiring expectations, portfolio strategy, networking, mentorship, and adapting creative skills across evolving industries and technologies.
Rather than a traditional lecture, this session is designed as an interactive conversation shaped by audience questions and discussion. Attendees will have the opportunity to engage directly with panelists on topics including career transitions, breaking into new industries, building sustainable creative careers, and navigating the rapidly evolving landscape of 3D tools and workflows.
Whether attendees are students, emerging professionals, or experienced artists exploring new directions, this session offers an opportunity to gain real-world insight from artists and technologists actively shaping the future of 3D across multiple industries.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Generative AI
Lighting
Full Conference Supporter
Full Conference
DescriptionEasyVFX is a resource-efficient framework for high-fidelity VFX synthesis. It leverages frequency-domain decomposition to disentangle spatial detail and motion dynamics, reducing learning complexity. A frequency-aware MoE enables efficient training, while a test-time adaptation strategy refines unseen effects. The approach achieves realistic, consistent results with minimal data and computation.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Generative AI
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionWe introduce a novel framework designed to generate continuous, coherent full-body motion from arbitrary audio streams with low latency. Our approach is a unified streaming architecture capable of synthesizing continuous motion from incremental audio inputs. And it serves as a plug-and-play solution for transforming voice agents into interactive humanoid avatars.
Art Paper
Arts & Design
Art
Computer Vision
Education
Fabrication
Generative AI
Geometry
Hardware
Image Processing
Math Foundations and Theory
Performance
Real-Time
Robotics
Full Conference Supporter
Full Conference
Experience
DescriptionEchoes of the Prior visualizes entropy within neural networks, turning machine forgetting into an interactive 4D experience. Unlike traditional CG simulating physical reality, this work reveals the aesthetics of algorithmic degradation, transforming technical failure into a new artistic medium bridging AI explainability with philosophical inquiry. It challenges the community to look beyond high-fidelity reconstruction and embrace the fragility of intelligence, proposing that data dissolution is as significant as its preservation.
Talks
Arts & Design
Gaming & Interactive
Production & Animation
Research & Education
Art
Computer Vision
Digital Twins
Education
Games
Image Processing
Pipeline Tools and Work
Rendering
Full Conference Supporter
Full Conference
DescriptionA defocused object viewed through the edge of another defocused object exhibits strange warping and reshaping effects, known as “eclipsed bokeh”. By studying what happens to the bokeh, we can see how to recreate the effect from 2D images.
Educator's Forum
Full Conference Supporter
Full Conference
Experience
DescriptionAfter four days of education-focused presentations, attendees are invited to an invigorating, open discussion on current topics in computer graphics education and interactive techniques. Come to the town hall session to ask questions, share perspectives, and have a good time with the SIGGRAPH education community!
Technical Workshop
New Technologies
Research & Education
Artificial Intelligence/Machine Learning
Computer Vision
Generative AI
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionThis workshop is dedicated to democratizing 3D content creation, encompassing both static and dynamic 3D assets, by exploring cutting-edge advances in volumetric video built on 4D Gaussian Splatting, fast 3D objects and scenes reconstruction, AI generated 3D content, and real-time immersive video generation with diffusion models.
We bring together both academic and industrial researchers and practitioners to discuss scalable pipelines that lower the barrier to immersive content creation. What's more, we will explore the 3D/4D content experience on both VR headsets as well as consumer mobile platforms.
A key feature of this workshop is a hands-on demonstration: participants will experience 1) volumetric video of industrial demonstration in both video-on-demand applications and in real-time streaming of
live-broadcasting, 2) interactive experience of AI generated 3D content and 3D reconstruction, and also 3) real-time interactive and high-resolution video generation via diffusion models.
The workshop seeks to bridge the gap between cutting-edge 3D content creation in academic research and in real-world industrial applications.
We bring together both academic and industrial researchers and practitioners to discuss scalable pipelines that lower the barrier to immersive content creation. What's more, we will explore the 3D/4D content experience on both VR headsets as well as consumer mobile platforms.
A key feature of this workshop is a hands-on demonstration: participants will experience 1) volumetric video of industrial demonstration in both video-on-demand applications and in real-time streaming of
live-broadcasting, 2) interactive experience of AI generated 3D content and 3D reconstruction, and also 3) real-time interactive and high-resolution video generation via diffusion models.
The workshop seeks to bridge the gap between cutting-edge 3D content creation in academic research and in real-world industrial applications.
Technical Paper
Research & Education
Animation
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionWe present a quadratic B-spline FEM approach for cloth simulation that captures smoother, more accurate deformations with rich wrinkle detail and few artifacts. Our optimized reduced integration scheme enables better performance compared to traditional linear FEM, while supporting robust, high-quality cloth simulation for realistic garment animation.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionWe introduce a hybrid framework for rendering dense hair and fur. By preserving fine geometric details for direct illumination while approximating complex multiple scattering within an equivalent anisotropic volume, our method achieves path-traced visual fidelity and soft glow of multiple scattering at a drastically reduced computational cost.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionWe extend the implicitly restarted Lanczos method to a multiscale context using arbitrary multigrids (algebraic, geometric) and we demonstrate the gain in performance and robustness on a variety of application scenarios.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Real-Time
Rendering
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Generative AI
Geometry
Modeling
Simulation
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionEgoForce enables real-time 3D hand pose, shape, and absolute position recovery from a single head-mounted RGB camera. Designed for lightweight smart glasses, it uses forearm cues and camera-aware ray-space lifting to provide precise tracking across diverse camera models and device setups.
Technical Paper
Research & Education
Animation
Fabrication
Haptics
Simulation
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionTo enable seamless mixed-reality telepresence in the wild, we present EgoRelight, a robust human performance capture and HDR illumination recovery approach using a single portable head-mounted device. This dual capability drives the photorealistic and relightable full-body 3D avatars, blending users naturally into surrounding environments.
Computer Animation Festival
Games Summit
Gaming & Interactive
Games
Full Conference Supporter
Full Conference
DescriptionIn an age long past, where gods and mortals coexisted in fragile harmony, Bjorn, a valiant Viking warrior, is entrusted with a sacred mission. Guided by the ancients, he must recover the fragments of the Astral Stone, a divine artifact shattered during the fall of the old realms. Across lands ravaged by chaos and haunted by forgotten creatures, Bjorn faces dark forces to restore balance to the worlds and save his tribe from a grim fate.
Art Paper
Arts & Design
Art
Computer Vision
Education
Fabrication
Generative AI
Geometry
Hardware
Image Processing
Math Foundations and Theory
Performance
Real-Time
Robotics
Full Conference Supporter
Full Conference
Experience
DescriptionElectrospun Fields advances SIGGRAPH’s interest in material computation by framing electric fields as programmable boundary conditions and fiber deposition as a material rendering of invisible forces. It contributes a reproducible UR20 robotic workflow for non-planar electrospinning, enabling ultra-light 3D membranes difficult to achieve with planar setups. By linking conductive scaffold topologies to material behaviors using bio-compatible polymers, this framework opens novel pathways for emergent and expressive computational art and fabrication.
Art Paper
Arts & Design
Art
Artificial Intelligence/Machine Learning
Audio
Augmented Reality
Digital Twins
Games
Geometry
Hardware
Performance
Real-Time
Scientific Visualization
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionAfter all the presentations, attendees are invited to participate in a Q&A.
Emerging Technologies
Arts & Design
Gaming & Interactive
New Technologies
Research & Education
Art
Artificial Intelligence/Machine Learning
Audio
Augmented Reality
Capture/Scanning
Computer Vision
Digital Twins
Display
Dynamics
Education
Fabrication
Games
Generative AI
Graphics Systems Architecture
Haptics
Hardware
Image Processing
Performance
Physical AI
Pipeline Tools and Work
Real-Time
Robotics
Scientific Visualization
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionEmerFlux is a two-layer liquid surface display that renders dynamic protruding pixels at the water–oil interface. By independently controlling an 11×11 pump array, it enables organic dot-based patterns and fluid aesthetic expressions directly on a living liquid surface.
Emerging Technologies
Technical Paper
Full Conference Supporter
Full Conference
Experience
DescriptionThis Emerging Technologies demo is a partner presentation to the Technical Paper:
A Generalizable Light Transport 3D Embedding for Global Illumination
We introduce a scalable 3D embedding that learns global illumination directly from 3D scene configurations. This design scales to environments with millions of triangles, enabling the first generalizable GI learning on complex, high-fidelity indoor scenes, far beyond prior limits. It bypasses per-scene retraining and screen-space artifacts to deliver realistic rendering.
Paper Presentation Presented: Tuesday, July 21, 2-3:30 pm; Paper Session: Light Transport; Room 408 A
A Generalizable Light Transport 3D Embedding for Global Illumination
We introduce a scalable 3D embedding that learns global illumination directly from 3D scene configurations. This design scales to environments with millions of triangles, enabling the first generalizable GI learning on complex, high-fidelity indoor scenes, far beyond prior limits. It bypasses per-scene retraining and screen-space artifacts to deliver realistic rendering.
Paper Presentation Presented: Tuesday, July 21, 2-3:30 pm; Paper Session: Light Transport; Room 408 A
Emerging Technologies
Technical Paper
Full Conference Supporter
Full Conference
Experience
DescriptionThis Emerging Technologies demo is a partner presentation to the Technical Paper:
A LoD of Gaussians: Out-of-Core Training and Rendering for Seamless Ultra-Large Scene Reconstruction
Gaussian Splatting struggles with large scenes due to memory limits and chunking artifacts. We introduce "A LoD of Gaussians," a new framework enabling ultra-large scene training and rendering on a single consumer GPU. Through out-of-core streaming and Level-of-Detail, it achieves seamless multi-scale reconstruction without relying on scene partitioning.
Paper Presentation Presented: Monday, July 20, 2-3:30 pm; Paper Session: 3D Gaussian Splatting I; Room 408 A
A LoD of Gaussians: Out-of-Core Training and Rendering for Seamless Ultra-Large Scene Reconstruction
Gaussian Splatting struggles with large scenes due to memory limits and chunking artifacts. We introduce "A LoD of Gaussians," a new framework enabling ultra-large scene training and rendering on a single consumer GPU. Through out-of-core streaming and Level-of-Detail, it achieves seamless multi-scale reconstruction without relying on scene partitioning.
Paper Presentation Presented: Monday, July 20, 2-3:30 pm; Paper Session: 3D Gaussian Splatting I; Room 408 A
Emerging Technologies
Technical Paper
Full Conference Supporter
Full Conference
Experience
DescriptionThis Emerging Technologies demo is a partner presentation to the Technical Paper:
Abstraction in Style: Beyond Texture and Color
Artistic styles often involve shape reinterpretation, not merely a change of surface appearance. Conventional style transfer methods miss this deeper abstraction behavior. We introduce Abstraction in Style (AiS), a two-stage generative framework that learns and transfers style at shape level, beyond texture and colors, enabling more expressive and controllable stylization.
Paper Presentation Presented: Thursday, July 23, 9-10:30 am; Paper Session: NPR, Style & Abstraction; Room 408 B
Abstraction in Style: Beyond Texture and Color
Artistic styles often involve shape reinterpretation, not merely a change of surface appearance. Conventional style transfer methods miss this deeper abstraction behavior. We introduce Abstraction in Style (AiS), a two-stage generative framework that learns and transfers style at shape level, beyond texture and colors, enabling more expressive and controllable stylization.
Paper Presentation Presented: Thursday, July 23, 9-10:30 am; Paper Session: NPR, Style & Abstraction; Room 408 B
Emerging Technologies
Technical Paper
Full Conference Supporter
Full Conference
Experience
DescriptionThis Emerging Technologies demo is a partner presentation to the Technical Paper:
AGILE: Hand-object Interaction Reconstruction from Video via Agentic Generation
AGILE reconstructs 3D hand-object interactions from monocular videos via agentic generation. It produces complete, simulation-ready object meshes and robustly estimates object pose without SfM, enabling accurate and stable results on challenging real-world data.
Paper Presentation Presented: Monday, July 20, 3:45-5:35 pm; Paper Session: Humans & Hands - Reconstruction; Room 403 A
AGILE: Hand-object Interaction Reconstruction from Video via Agentic Generation
AGILE reconstructs 3D hand-object interactions from monocular videos via agentic generation. It produces complete, simulation-ready object meshes and robustly estimates object pose without SfM, enabling accurate and stable results on challenging real-world data.
Paper Presentation Presented: Monday, July 20, 3:45-5:35 pm; Paper Session: Humans & Hands - Reconstruction; Room 403 A
Emerging Technologies
Technical Paper
Full Conference Supporter
Full Conference
Experience
DescriptionThis Emerging Technologies demo is a partner presentation to the Technical Paper:
Controllable Texture Tiling via Diffusion Transformers with Transformed Rotary Embeddings
We introduce a novel Diffusion Transformer framework for high-fidelity, controllable texture tiling. By leveraging Coordinate-Transformed Rotary Embeddings, our method enables precise manipulation of texture frequency, scale, and orientation. It seamlessly integrates reference patterns into images while perfectly preserving the original scene's lighting and geometry.
Paper Presentation Presented: Tuesday, July 21, 10:45 am-12:15 pm; Paper Session: Materials & Textures; Room 408 A
Controllable Texture Tiling via Diffusion Transformers with Transformed Rotary Embeddings
We introduce a novel Diffusion Transformer framework for high-fidelity, controllable texture tiling. By leveraging Coordinate-Transformed Rotary Embeddings, our method enables precise manipulation of texture frequency, scale, and orientation. It seamlessly integrates reference patterns into images while perfectly preserving the original scene's lighting and geometry.
Paper Presentation Presented: Tuesday, July 21, 10:45 am-12:15 pm; Paper Session: Materials & Textures; Room 408 A
Emerging Technologies
Technical Paper
Full Conference Supporter
Full Conference
Experience
DescriptionThis Emerging Technologies demo is a partner presentation to the Technical Paper:
EgoForce: Forearm-Guided Camera-Space 3D Hand Pose from a Monocular Egocentric Camera
EgoForce enables real-time 3D hand pose, shape, and absolute position recovery from a single head-mounted RGB camera. Designed for lightweight smart glasses, it uses forearm cues and camera-aware ray-space lifting to provide precise tracking across diverse camera models and device setups.
Paper Presentation Presented: Monday, July 20, 3:45-5:35 pm; Paper Session: Humans & Hands - Reconstruction; Room 403A
EgoForce: Forearm-Guided Camera-Space 3D Hand Pose from a Monocular Egocentric Camera
EgoForce enables real-time 3D hand pose, shape, and absolute position recovery from a single head-mounted RGB camera. Designed for lightweight smart glasses, it uses forearm cues and camera-aware ray-space lifting to provide precise tracking across diverse camera models and device setups.
Paper Presentation Presented: Monday, July 20, 3:45-5:35 pm; Paper Session: Humans & Hands - Reconstruction; Room 403A
Emerging Technologies
Technical Paper
Full Conference Supporter
Full Conference
Experience
DescriptionThis Emerging Technologies demo is a partner presentation to the Technical Paper:
GauSmoke: Hybrid Physics-Optical Gaussian Splatting for Sparse Smoke Reconstruction
We present a physics-aware method for reconstructing dynamic fluids from sparse-view videos. By integrating volumetric rendering with physically guided Gaussian optimization, it enforces consistency in density and motion, reducing artifacts and improving realism. The approach achieves accurate, stable 3D smoke reconstruction with high visual and physical fidelity.
Paper Presentation Presented: Tuesday, July 21, 9-10:30 am; Paper Session: Fluid Reconstruction & Optimization; Room 408 A
GauSmoke: Hybrid Physics-Optical Gaussian Splatting for Sparse Smoke Reconstruction
We present a physics-aware method for reconstructing dynamic fluids from sparse-view videos. By integrating volumetric rendering with physically guided Gaussian optimization, it enforces consistency in density and motion, reducing artifacts and improving realism. The approach achieves accurate, stable 3D smoke reconstruction with high visual and physical fidelity.
Paper Presentation Presented: Tuesday, July 21, 9-10:30 am; Paper Session: Fluid Reconstruction & Optimization; Room 408 A
Emerging Technologies
Technical Paper
Full Conference Supporter
Full Conference
Experience
DescriptionThis Emerging Technologies demo is a partner presentation to the Technical Paper:
LightOpt: Lights Optimization for Real-time Rendering
LightOpt is a differentiable optimization framework that automatically reduces the number of lights in real-time game scenes while preserving visual appearance. By optimizing and restructuring real-time light sources, it lowers lighting cost, reduces overlap, and improves cross-platform performance for modern game rendering.
Paper Presentation Presented:
Monday, July 20, 3:45-5:15 pm; Paper Session: Real-Time Rendering; Room 408 A
LightOpt: Lights Optimization for Real-time Rendering
LightOpt is a differentiable optimization framework that automatically reduces the number of lights in real-time game scenes while preserving visual appearance. By optimizing and restructuring real-time light sources, it lowers lighting cost, reduces overlap, and improves cross-platform performance for modern game rendering.
Paper Presentation Presented:
Monday, July 20, 3:45-5:15 pm; Paper Session: Real-Time Rendering; Room 408 A
Emerging Technologies
Technical Paper
Full Conference Supporter
Full Conference
Experience
DescriptionThis Emerging Technologies demo is a partner presentation to the Technical Paper:
LooseRoPE: Content-aware Attention Manipulation for Semantic Harmonization
LooseRoPE enables prompt-free image editing by letting users crop and paste objects directly into new scenes. It improves diffusion-based harmonization by modulating RoPE with saliency, balancing identity preservation and contextual blending. This yields intuitive, spatially precise edits that maintain object appearance, pose, and placement without requiring text prompts or masks.
Paper Presentation Presented: Thursday, July 23, 10:45 am-12:15 pm; Paper Session: Image Editing,; Room 408 B
LooseRoPE: Content-aware Attention Manipulation for Semantic Harmonization
LooseRoPE enables prompt-free image editing by letting users crop and paste objects directly into new scenes. It improves diffusion-based harmonization by modulating RoPE with saliency, balancing identity preservation and contextual blending. This yields intuitive, spatially precise edits that maintain object appearance, pose, and placement without requiring text prompts or masks.
Paper Presentation Presented: Thursday, July 23, 10:45 am-12:15 pm; Paper Session: Image Editing,; Room 408 B
Emerging Technologies
Technical Paper
Full Conference Supporter
Full Conference
Experience
DescriptionThis Emerging Technologies demo is a partner presentation to the Technical Paper:
MotionBricks: Scalable Real-Time Motions with Modular Latent Generative Model and Smart Primitives
MotionBricks is a real-time generative framework that transforms interactive motion control for animation and robotics. By combining a large-scale latent backbone with intuitive "smart primitives," it delivers high-quality, zero-shot motion synthesis at 15,000 FPS, allowing users to effortlessly build complex and fine-grained animations and robotic applications.
Paper Presentation Presented: Tuesday, July 21, 2-3:30 pm; Paper Session: Motion Generation, Warping & Control; Room 403 A
MotionBricks: Scalable Real-Time Motions with Modular Latent Generative Model and Smart Primitives
MotionBricks is a real-time generative framework that transforms interactive motion control for animation and robotics. By combining a large-scale latent backbone with intuitive "smart primitives," it delivers high-quality, zero-shot motion synthesis at 15,000 FPS, allowing users to effortlessly build complex and fine-grained animations and robotic applications.
Paper Presentation Presented: Tuesday, July 21, 2-3:30 pm; Paper Session: Motion Generation, Warping & Control; Room 403 A
Emerging Technologies
Technical Paper
Full Conference Supporter
Full Conference
Experience
DescriptionThis Emerging Technologies demo is a partner presentation to the Technical Paper:
MUSIC: Learning Muscle-Driven Dexterous Hand Control
We present a data-driven framework for physics-based, muscle-driven dexterous control that enables musculoskeletal hands to perform piano playing for novel musical scores. Our framework adopts a hierarchical architecture, combining high-frequency muscle control with low-frequency coordination. The method achieves coordinated bimanual motions and produces human-like activation patterns, generalizing beyond reference data.
Paper Presentation Presented: Thursday, July 23, 2-3:30 pm; Paper Session: Learning in Motion; Room 403 A
MUSIC: Learning Muscle-Driven Dexterous Hand Control
We present a data-driven framework for physics-based, muscle-driven dexterous control that enables musculoskeletal hands to perform piano playing for novel musical scores. Our framework adopts a hierarchical architecture, combining high-frequency muscle control with low-frequency coordination. The method achieves coordinated bimanual motions and produces human-like activation patterns, generalizing beyond reference data.
Paper Presentation Presented: Thursday, July 23, 2-3:30 pm; Paper Session: Learning in Motion; Room 403 A
Emerging Technologies
Technical Paper
Full Conference Supporter
Full Conference
Experience
DescriptionThis Emerging Technologies demo is a partner presentation to the Technical Paper:
Sketch2Arti: Sketch-based Articulation Modeling of CAD Objects
Sketch2Arti is the first sketch-based articulation modeling system for CAD objects. Given a CAD model and simple 2D sketches, it identifies movable parts and predicts motion parameters. It can also complete missing internal structures revealed by motion, enabling controllable, iterative, and generalizable articulation modeling for diverse objects.
Paper Presentation Presented: Wednesday, July 22, 10:45 am-12:35 pm; Paper Session: Vectors Graphics, Strokes & Sketches; Room 403 A
Sketch2Arti: Sketch-based Articulation Modeling of CAD Objects
Sketch2Arti is the first sketch-based articulation modeling system for CAD objects. Given a CAD model and simple 2D sketches, it identifies movable parts and predicts motion parameters. It can also complete missing internal structures revealed by motion, enabling controllable, iterative, and generalizable articulation modeling for diverse objects.
Paper Presentation Presented: Wednesday, July 22, 10:45 am-12:35 pm; Paper Session: Vectors Graphics, Strokes & Sketches; Room 403 A
Emerging Technologies
Arts & Design
Gaming & Interactive
New Technologies
Research & Education
Art
Artificial Intelligence/Machine Learning
Audio
Augmented Reality
Capture/Scanning
Computer Vision
Digital Twins
Display
Dynamics
Education
Fabrication
Games
Generative AI
Graphics Systems Architecture
Haptics
Hardware
Image Processing
Performance
Physical AI
Pipeline Tools and Work
Real-Time
Robotics
Scientific Visualization
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionEmoMime is a neck-mounted wearable robotic system designed to augment social behavior and self-expression. Utilizing a soft-rigid hybrid structure, it enables safe, expressive interactions near the user's face. We invite attendees to experience how a human-like robotic hand near the face dynamically amplifies social signals and enables comforting self-touch.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Generative AI
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionPreserving lip synchronization during emotion manipulation in video is a major challenge. This paper proposes a method using a pretrained StyleGAN and a latent-based landmark editing module. It modifies the latent edit direction with facial landmarks, enabling fast, high-quality emotion changes while sustaining original lip sync.
Emerging Technologies
Arts & Design
Gaming & Interactive
New Technologies
Research & Education
Art
Artificial Intelligence/Machine Learning
Audio
Augmented Reality
Capture/Scanning
Computer Vision
Digital Twins
Display
Dynamics
Education
Fabrication
Games
Generative AI
Graphics Systems Architecture
Haptics
Hardware
Image Processing
Performance
Physical AI
Pipeline Tools and Work
Real-Time
Robotics
Scientific Visualization
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionEnableController democratizes gaming accessibility by transforming the human body into a universal input device. Using Generative AI to translate natural language into kinematic controls, and virtual emulation for compatibility, it allows users with disabilities to play any commercial game using therapeutic or comfortable body movements—no coding required.
Educator's Forum
Arts & Design
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Animation
Art
Computer Vision
Digital Twins
Education
Games
Modeling
Pipeline Tools and Work
Rendering
Simulation
Spatial Computing
Full Conference Supporter
Full Conference
Experience
DescriptionLimbitless ENGAGE dismantles educational silos through human-centric bionics. This scalable STEAM framework integrates digital arts, game design, and engineering via "near-peer" mentorship. Students utilize a replicable "Bionic Blueprint" to master graphics for social impact, empowering the next generation to design for humanity through empathy-driven, interdisciplinary collaboration.
Immersive Pavilion
Arts & Design
Gaming & Interactive
New Technologies
Research & Education
Animation
Art
Artificial Intelligence/Machine Learning
Audio
Augmented Reality
Computer Vision
Digital Twins
Education
Ethics and Society
Fabrication
Games
Generative AI
Haptics
Hardware
Image Processing
Modeling
Performance
Pipeline Tools and Work
Real-Time
Scientific Visualization
Simulation
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionThis study introduces a real-time feedback system for VR exergames. By leveraging LLMs to analyze personal movement data against historical datasets , the system provides personalized voice encouragement to support reflection-in-action and aim to enhance players' performance.
Appy Hour
Arts & Design
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Animation
Art
Artificial Intelligence/Machine Learning
Audio
Augmented Reality
Computer Vision
Digital Twins
Education
Ethics and Society
Games
Generative AI
Geometry
Image Processing
Math Foundations and Theory
Modeling
Performance
Physical AI
Pipeline Tools and Work
Real-Time
Robotics
Scientific Visualization
Simulation
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionEnvoxel is an interactive voxel environment and structure generation app. It turns prompts into editable block structures, preserving interiors and materials while streaming predictions live in the browser. The demo shows upload, cut, completion, side-by-side inspection, and export for game-native voxel worlds.
Talks
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Animation
Dynamics
Games
Generative AI
Geometry
Modeling
Performance
Pipeline Tools and Work
Real-Time
Simulation
Full Conference Supporter
Full Conference
DescriptionFramestore introduces the Anatomy Toolkit (FAT), replacing legacy deformer workflows with an end-to-end simulation platform. FAT automates constraint construction through geometric labeling and enables the convergence of Rigging and Creature FX workflows. This delivers anatomically accurate, high-fidelity muscle and tissue deformations with superior efficiency and simplified artist experience.
Technical Paper
Research & Education
Geometry
Modeling
Real-Time
Rendering
Full Conference Supporter
Full Conference
DescriptionThis article introduces an algorithm that exactly constructs the so-called Weiler model (also called a 3D mesh arrangement) and that uses it to implement CSG with arbitrary multi-operand expressions. The main contribution is a 2D Constrained Delaunay Triangulation with exact coordinates and symbolic perturbations to disambiguate configurations with co-cyclic points.
Computer Animation Festival
Full Conference Supporter
Full Conference
DescriptionNorah, a seasoned war correspondent, is living with cancer. Determined to make the most of the time she has left, she turns to Nadim, her closest friend and former cameraman, to join her for one final assignment.
Art Paper
Arts & Design
Art
Artificial Intelligence/Machine Learning
Audio
Augmented Reality
Digital Twins
Games
Geometry
Hardware
Performance
Real-Time
Scientific Visualization
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionThis work treats AI models not only as technical systems, but as perceptual, embodied, and cultural artifacts. By combining voice cloning and latent space visualization in an interactive installation, the project demonstrates how AI representations can become accessible and critically examined. It opens new perspectives on translation, embodiment, and authorship, offering artists, researchers, and technologists a shared framework to reflect on how computational representations shape creative practice and human identity.
Emerging Technologies
Arts & Design
Gaming & Interactive
New Technologies
Research & Education
Art
Artificial Intelligence/Machine Learning
Audio
Augmented Reality
Capture/Scanning
Computer Vision
Digital Twins
Display
Dynamics
Education
Fabrication
Games
Generative AI
Graphics Systems Architecture
Haptics
Hardware
Image Processing
Performance
Physical AI
Pipeline Tools and Work
Real-Time
Robotics
Scientific Visualization
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionExperience Sharing Studio (ESS) lets visitors relive recorded human experiences through synchronized spatial video, sound, haptics, floor feedback, and 3D poses. Using the Experium Platform and unique Experience IDs, ESS connects audiovisual and embodied information into interactive playback. Visitors can explore both dynamic actions and quiet moments at their own pace through the Spatio-Temporal Progress Bar and User-Driven Playback Mode.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Full Conference Supporter
Full Conference
DescriptionOur paper presents a method for shape-morphing and deformation using an exact mathematical flow. We introduce a neural architecture guaranteed to encode a flow, leveraging implicit supervision of explicit flows, ensuring topological consistency and efficient forward/inverse computation without ODE solvers. Applications include shape-morphing, editing, and transitions via a reference shape.
Immersive Pavilion
Arts & Design
Gaming & Interactive
New Technologies
Research & Education
Animation
Art
Artificial Intelligence/Machine Learning
Audio
Augmented Reality
Computer Vision
Digital Twins
Education
Ethics and Society
Fabrication
Games
Generative AI
Haptics
Hardware
Image Processing
Modeling
Performance
Pipeline Tools and Work
Real-Time
Scientific Visualization
Simulation
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionA neuro-interactive installation merging cinema and neuroscience. Using EEG and gaze tracking, this exhibition monitors the audience's live attentional shifts during implicit narrative choices. Participants can gain insights into their own brain states as they transition from passive observers to active agents as they engage in an implicitly interactive narrative experience.
Talks
Arts & Design
New Technologies
Production & Animation
Research & Education
Art
Artificial Intelligence/Machine Learning
Games
Generative AI
Industry Insight
Lighting
Modeling
Physical AI
Pipeline Tools and Work
Rendering
Scientific Visualization
Simulation
Full Conference Supporter
Full Conference
DescriptionThis talk introduces a novel framework that bridges ray tracing and quantum physics. We use Quantum Collision Models to simulate materials with optical properties that dynamically evolve in response to light. This enables the synthesis of new materials with complex optical signatures, suitable for production pipelines and near-term quantum hardware.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Fabrication
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Generative AI
Modeling
Full Conference Supporter
Full Conference
DescriptionWe study fabric try-on as a garment re-texturing task, replacing fabric materials while preserving garment geometry and illumination. Leveraging modern image editing models and a real-image data curation pipeline, our framework performs fabric removal and application directly in image space, enabling coherent and photorealistic fabric try-on from a single image.
Talks
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Animation
Dynamics
Games
Generative AI
Geometry
Modeling
Performance
Pipeline Tools and Work
Real-Time
Simulation
Full Conference Supporter
Full Conference
DescriptionWe describe a new rigging technique that generates face parametrizations by leveraging curvenets as a sparse proxy for authoring weights outlining face contours, which are then interpolated across the surface mesh to drive face articulation. We showcase our method using face rigs built for various characters from Disney/Pixar’s Hoppers.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Generative AI
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Modeling
Real-Time
Rendering
Full Conference Supporter
Full Conference
DescriptionWe propose a BRDF modeling method that provides accurate and compact representations for isotropic and anisotropic BRDFs. We propose new theoretical developments enabling tractable Gaussian Process BRDF regression, leading to analytical BRDF representations. State-of-the-art methods can be outperformed in most BRDF cases by using a small training set of observations.
Technical Paper
Research & Education
Fabrication
Geometry
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionGeneralized winding numbers are a useful measure for inside-outside classification. This paper introduces a novel method for computing winding numbers for triangle meshes, which leverages a ray-mesh intersection and an elementary sum over boundaries. Our algorithm gives the exact answer and is more efficient than all existing techniques in practice.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Fabrication
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionWe present a fast sparse matrix permutation algorithm tailored to mesh-based direct solvers. By exploiting mesh structure through patch-based nested dissection, our method reduces permutation overhead while preserving high-quality orderings, leading to substantial end-to-end speedups in sparse Cholesky factorization on both CPUs and GPUs.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Real-Time
Rendering
Scientific Visualization
Simulation
Full Conference Supporter
Full Conference
DescriptionFastVEM is an efficient boundary-conforming fluid simulation framework. It combines Virtual Element discretization, a simulation-friendly body-fitted grid construction strategy, and a tailored geometric multigrid method to achieve robust, high-fidelity, and efficient fluid–boundary interaction.
Talks
Production & Animation
Animation
Pipeline Tools and Work
Real-Time
Full Conference Supporter
Full Conference
DescriptionWe present a scalable workflow for importing and managing film production environment assets. By using Unreal Engine and OpenUSD, this workflow reduces import times, load times, and data footprints while improving framerate. This avoids traditional DCC geometry overhead and allows teams to rapidly import polygon-dense assets with increased interactive performance.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Real-Time
Rendering
Full Conference Supporter
Full Conference
DescriptionWe present a structure-aware densification method for 3D Gaussian Splatting that adapts primitives based on local texture frequency. By combining multi-scale analysis and anisotropic splitting, it preserves fine details while avoiding unnecessary density, enabling faster training and higher-quality novel view synthesis.
Technical Paper
Research & Education
Geometry
Modeling
Real-Time
Rendering
Full Conference Supporter
Full Conference
DescriptionThis paper introduces a novel method for generating offset meshes from 3D surfaces. It uniquely supports mitered joins and variable offset distances while robustly preserving sharp features. Evaluated on a broad dataset, our approach outperforms state-of-the-art methods in feature fidelity and element count.
Immersive Pavilion
Arts & Design
Gaming & Interactive
New Technologies
Research & Education
Animation
Art
Artificial Intelligence/Machine Learning
Audio
Augmented Reality
Computer Vision
Digital Twins
Education
Ethics and Society
Fabrication
Games
Generative AI
Haptics
Hardware
Image Processing
Modeling
Performance
Pipeline Tools and Work
Real-Time
Scientific Visualization
Simulation
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionFeltSight is a wearable mixed reality haptic experience that shifts perception from sight to touch. A haptic glove renders audio-driven vibrations as textures, while a pared-back XR display shows point-cloud objects that appear only when explored. Inspired by star-nosed moles, it promotes meditative, more-than-human embodied sensing.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Generative AI
Lighting
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Technical Paper
Research & Education
Display
Generative AI
Hardware
Image Processing
Full Conference Supporter
Full Conference
DescriptionImage and audio processing applications want to use large FIR filters while adhering to strict runtime and latency requirements. Prior methods for fast filter approximation offer limited speed versus quality tradeoffs. We unify these approximation techniques as primitives within an automatically searchable space for fast 1D and 2D filters.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionWe introduce FIT (Fit-Inclusive Try-on), a virtual try-on dataset of over 1M samples covering diverse garment fits, each annotated with precise body/garment measurements. Our novel data generation pipeline leverages synthetic garment simulation and photorealistic retexturing. Then, we leverage FIT to train a state-of-the-art fit-aware virtual try-on model.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Generative AI
Geometry
Modeling
Simulation
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionWe propose FLASHand, the first feed-forward model to reconstruct high-fidelity, relightable, and animatable 3D hand avatars from a single RGB image. By leveraging NIMBLE priors and mesh-based 2D Gaussian Splatting, FLASHand achieves instant, personalized reconstruction with disentangled appearance, supporting real-time animation and relighting.
Spatial Storytelling
Arts & Design
Production & Animation
Research & Education
Capture/Scanning
Image Processing
Rendering
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionWe present Flashover, a spatial storytelling project that reframes wildfire as lived experience: memory, repetition, and loss. We describe how multi-zone installation design and point-based volumetric imagery invite active perception rather than spectacle, translating traumatic memory into embodied encounter and reflection on the climate crisis.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Full Conference Supporter
Full Conference
DescriptionGenerating CAD B-Reps is challenging due to complex topology-geometry entanglement. We solve this by reformulating B-Reps into compositional k-cell particles. Our flow matching framework unifies topological entities, enabling the joint generation, precise 3D reconstruction, and local in-painting of high-fidelity CAD models with superior validity.
Technical Paper
Research & Education
Animation
Geometry
Hardware
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionWe present solutions to the floating-point–induced decision errors arising in CCD for parametric surfaces, including an error-resistant CCD method and a data construction pipeline with known GT for algorithm evaluation under floating-point perturbations. Experimental results demonstrate that the proposed methods are effective and robust.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Real-Time
Rendering
Scientific Visualization
Simulation
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Course
New Technologies
Production & Animation
Research & Education
Dynamics
Games
Simulation
Full Conference Supporter
Full Conference
DescriptionLattice Boltzmann Method (LBM) offers a powerful alternative to traditional Navier-Stokes solvers, using mesoscopic particle distributions on Cartesian lattices to achieve far superior GPU performance and accuracy. This 1.5h course introduces LBM fundamentals and recent advances in multiphase flows, free-surface flows, high–Reynolds-number flows, moving boundaries, and robotics. We also highlight applications of LBM in industrial simulation, game production, and visual effects.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Real-Time
Rendering
Full Conference Supporter
Full Conference
DescriptionWe introduce an end-to-end adaptive sampling and denoising pipeline for sparse real-time path tracing. Our method trains stably despite discrete sampling decisions and uses perceptual, tonemapping-aware optimization to place samples where they matter most. It samples fine structures, specular highlights, shadows, producing more detailed reconstructions than superresolution or uniform sampling.
Games Summit
Gaming & Interactive
Games
Full Conference Supporter
Full Conference
Experience
DescriptionAs the game industry grapples with rising costs, shifting platforms, and an explosion of available tools, the question of how studios build their games has never been more consequential. IO Interactive Technical Director Henrik Schlichter sits down to discuss, debate, and unpack the realities of modern game development - what it takes to build the custom Glacier engine, innovate at scale, and how to build cheaper, faster and better without sacrificing quality. Equal parts technical deep-dive and industry reality check, this fireside chat sets the stage for bold conversations for Games at SIGGRAPH 2026.
Technical Paper
Research & Education
Fabrication
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionWe propose a method to approximate free-form surfaces using rotational patches for architectural rationalization. By segmenting the input surface and optimizing B-spline-based patch layouts, followed by post-processing, we generate smooth, seamless quad meshes with locally repeated building elements (including beams, panels, and nodes), offering great potential for reducing fabrication costs.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Generative AI
Full Conference Supporter
Full Conference
DescriptionFreeOrbit4D is a training-free framework for camera redirection from monocular videos. By decoupling foreground and background reconstruction and completing foreground geometry via multi-view diffusion, it builds a foreground-complete 4D proxy that guides video generation, enabling faithful and temporally coherent novel views under large-angle camera trajectories.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Fabrication
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionFreeShell introduces a robust thermal-shrinkage-actuated 4D printing technique for fabricating freeform thin-shell surfaces. By printing triangular tiles connected by shrinkable connectors using a single material, heating triggers the transformation from flat structures into 3D shells. An optimized mesh layout algorithm ensures geometric accuracy and robustness, reducing material and environment dependency.
Talks
Arts & Design
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Animation
Art
Artificial Intelligence/Machine Learning
Ethics and Society
Generative AI
Industry Insight
Pipeline Tools and Work
Full Conference Supporter
Full Conference
DescriptionProfessional GenAI production lacks auditable decision-making, failing legal scrutiny. Shifting from authenticity to "defensibility," Verifiable Story Production (VSP) operationalizes responsibility via Phase 0 (Human Intent) and four pillars: Sovereign Ingest, Constitutional Guardrails, Persistent Provenance, and Human Validation. VSP is an open standard now being evaluated in film school projects.
Art Paper
Arts & Design
Animation
Art
Artificial Intelligence/Machine Learning
Audio
Capture/Scanning
Computer Vision
Display
Ethics and Society
Generative AI
Hardware
Image Processing
Physical AI
Real-Time
Robotics
Scientific Visualization
Simulation
Spatial Computing
Full Conference Supporter
Full Conference
Experience
DescriptionThe artwork HUMAN OVERS[A]IGHT: THE OPS ROOM is relevant to the SIGGRAPH community as it introduces the critical issues of lawmaking in the field of Artificial Intelligence. By bridging these disciplines, this translation aims to open a broader discourse on the topic from an innovative standpoint in media art that integrates legal matters, computer science, and a critical media theory approach.
Technical Workshop
Research & Education
Art
Artificial Intelligence/Machine Learning
Augmented Reality
Computer Vision
Display
Image Processing
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionVisual computing encompasses the full range of technologies and methodologies involved in rendering, displaying, reconstructing, modeling, and recording visual data. Among the many subfields within visual computing, there has been a growing interest in two subfields in particular: human vision and machine vision. Human vision research focuses on understanding how human vision and cognition functions and using that knowledge to improve the user experience through realism, immersion, and responsiveness. In contrast, machine vision research focuses on building computational systems that complete a visual perception task, and may aim to mimic the structure or functionality of biological vision systems.
This workshop aims to facilitate dialogue and collaboration between researchers and practitioners in machine and human vision. By exploring common challenges, methodologies, and philosophical approaches, ranging from perceptually grounded models to data-driven deep learning frameworks, we hope to spark interdisciplinary discussion that drives innovation across the broader field of visual computing.
This workshop aims to facilitate dialogue and collaboration between researchers and practitioners in machine and human vision. By exploring common challenges, methodologies, and philosophical approaches, ranging from perceptually grounded models to data-driven deep learning frameworks, we hope to spark interdisciplinary discussion that drives innovation across the broader field of visual computing.
Talks
Production & Animation
Research & Education
Animation
Art
Capture/Scanning
Digital Twins
Dynamics
Industry Insight
Lighting
Pipeline Tools and Work
Rendering
Simulation
Full Conference Supporter
Full Conference
DescriptionIn the film Hoppers, a forest fire rages and engulfs the glade which Mabel swore to protect. We describe how we crafted this fire through our production strategy, looks and rig development, fire ‘taxonomy’ and dressing, lighting and compositing workflows, and optimizations such as an optimized fire lighting representation.
Educator's Forum
Arts & Design
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Animation
Art
Computer Vision
Digital Twins
Education
Games
Modeling
Pipeline Tools and Work
Rendering
Simulation
Spatial Computing
Full Conference Supporter
Full Conference
Experience
DescriptionAs the tools of photography and computer graphics become inseparable, the 'hybrid artist' is the new industry standard. This talk demonstrates how to bridge the lens-to-render gap by teaching students to navigate the collaborative methodologies of modern VFX and commercial production."
Frontiers
Research & Education
Education
Full Conference Supporter
Full Conference
Experience
DescriptionWe recently recovered UNIX V4 from a 1974 magnetic tape at the University of Utah. This version of the UNIX operating system, thought to have been lost, was the 19th copy distributed to the public, just months after the first public announcement. It was originally acquired by Martin Newell while managing the computer graphics laboratory, and it was likely connected to his foundational research in procedural modeling and the famous Utah teapot. This talk shows the unique insight this artifact provides into the dawn of computer graphics and the history of UNIX.
Frontiers
Arts & Design
Research & Education
Art
Education
Ethics and Society
Full Conference Supporter
Full Conference
Experience
DescriptionMost programmers don't want people looking too closely at their code, except the parts where they found that clever hack or ingenious workaround. And now with AI some people don't even read the code they generate. But what if we looked closer at code? This talk introduces Critical Code Studies, a field that uses computer source code as an entrypoint for discussing culture, philosophy, and the way we see and represent the world. Focusing on examples from the world of computer graphics, I will explore what your code may say about you and about the world you are modeling and envisioning.
Frontiers
New Technologies
Digital Twins
Physical AI
Robotics
Scientific Visualization
Simulation
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionThis workshop addresses the transformative role of digital twins across science and industry. Following the National Academies' 2024 standardized definition, we seek to foster a community of practice embracing this common understanding across diverse domains. We open with a 2026 state-of-the-art survey and Digital Twin Consortium perspectives on evolving standards, then examine societal impacts in healthcare, fusion energy, climate science, and sovereign AI. Technical sessions address rendering fidelity in virtual replicas, AI integration for prediction and optimization, and frameworks combining machine learning with physics-based simulation. Case studies span drug discovery, computational biomedicine, sustainable energy, entertainment production, and scientific visualization.
Frontiers
Frontier Workshop: Dronevision, Holodecks, and Spatial Computing Using Swarms of Flying Light Specks
9:00am - 12:15pm PDT Thursday, 23 July 2026 Room 501 ABCArts & Design
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Animation
Art
Digital Twins
Display
Geometry
Graphics Systems Architecture
Haptics
Hardware
Networking
Performance
Physical AI
Real-Time
Robotics
Scientific Visualization
Spatial Computing
Full Conference Supporter
Full Conference
Experience
DescriptionThis timely SIGGRAPH 2026 Frontiers workshop explores Flying Light Specks (FLSs), including their applications, current capabilities, and future research directions. An FLS is a miniature drone equipped with a lighting apparatus that serves as a building block for next-generation display systems. These systems range from desktop Dronevision platforms to room-scale holodecks and outdoor spatial-computing environments. Swarms of FLSs can illuminate graphics and animations visible to the naked eye, enabling users to interact with them using their bare hands. Topics include the classification of FLS devices with LightBender as a representative example, end-to-end architectures spanning content generation, illumination, and bare-hand interaction.
For more information visit: https://www.flsworkshop.com
For more information visit: https://www.flsworkshop.com
Frontiers
Arts & Design
New Technologies
Research & Education
Animation
Art
Display
Fabrication
Image Processing
Full Conference Supporter
Full Conference
Experience
DescriptionThis frontiers workshop equips participants with hands-on experience using a modern computational design workflow for creating and fabricating barrier-grid animations. Participants will author animations whose occlusion patterns are shaped by mathematical functions, preview them under multiple interaction modes, and fabricate take-home artifacts. By participating in this workshop, attendees will understand how to design expressive barrier-grid animations that balance form and function.
Frontiers
New Technologies
Research & Education
Artificial Intelligence/Machine Learning
Augmented Reality
Capture/Scanning
Computer Vision
Digital Twins
Industry Insight
Real-Time
Robotics
Scientific Visualization
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionWe're back for round 2 of "From the Dev Floor to the Operating Room"! Like last year, we hope to bring to light more of the wonderful applications of computer graphics in medicine. We'll be discussing current challenges with XR in medicine, digital twins, simulation, robotics and more.
We'll also expand on last years workshop activity with a brand new panel of physicians to help us practice designing graphics solutions for medicine that actually make an impact.
This will also be a great place to network with people interested the beautiful intersection of Medicine and Technology
We'll also expand on last years workshop activity with a brand new panel of physicians to help us practice designing graphics solutions for medicine that actually make an impact.
This will also be a great place to network with people interested the beautiful intersection of Medicine and Technology
Frontiers
Research & Education
Artificial Intelligence/Machine Learning
Full Conference Supporter
Full Conference
Experience
DescriptionQuantum Machine Learning (QML) is emerging as a transformative field connecting quantum computing with modern machine learning and AI. This workshop explores how quantum-based approaches can support new learning paradigms, hybrid quantum–classical architectures, generative models, and computer graphics pipelines. Emphasizing both theory and practice, the workshop will examine near-term quantum hardware, deployable system design, and rigorous evaluation methods for identifying meaningful quantum advantage. Through three keynote talks, a panel discussion, open dialogue, and roadmap synthesis, the workshop aims to foster interdisciplinary collaboration and help shape the future of scalable, practical, and impactful QML.
Technical Paper
Research & Education
Games
Real-Time
Rendering
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionWe propose using Gabor kernels as volumetric density primitives to enable compact, cheap and continuous level-of-detail to primitive-based physically-based volume rendering with single and multiple scattering. We show state-of-the-art results in regression quality, runtime performance and introduce novel stochastic techniques for trading quality and performance in volume rendering.
Games Summit
Gaming & Interactive
Audio
Games
Full Conference Supporter
Full Conference
Experience
DescriptionJoin a Composer and Sound Designer as they share a glimpse into how they build sonic palettes and craft unique audio experiences for games. Hear examples of how various layers of different sounds can create a fireball spell, which unique combination of instruments create an unsettling atmosphere, or even how to evoke a specific emotion within just a few seconds of audio.
In addition to diving into the details on how these sound effects and musical excerpts were made, we’ll also focus on the high level decisions that would inform the audio’s direction. Analyzing how to determine that which is most important for audio to represent, aligning across multiple teams on creative goals, and other essential considerations for the production of audio for games.
In addition to diving into the details on how these sound effects and musical excerpts were made, we’ll also focus on the high level decisions that would inform the audio’s direction. Analyzing how to determine that which is most important for audio to represent, aligning across multiple teams on creative goals, and other essential considerations for the production of audio for games.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionGarment Particles is a 5D point-cloud representation that jointly encodes 2D sewing patterns and 3D geometry. This representation enables Garment Particles Flow (GPF) and Particles-to-Pattern Flow (PPF), a rectified flow framework for both simulation-ready sewing pattern generation from text and images and intuitive 2D/3D editing via diffusion posterior sampling.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Real-Time
Rendering
Scientific Visualization
Simulation
Full Conference Supporter
Full Conference
DescriptionWe present a physics-aware method for reconstructing dynamic fluids from sparse-view videos. By integrating volumetric rendering with physically guided Gaussian optimization, it enforces consistency in density and motion, reducing artifacts and improving realism. The approach achieves accurate, stable 3D smoke reconstruction with high visual and physical fidelity.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Real-Time
Rendering
Full Conference Supporter
Full Conference
DescriptionWe propose Gaussian point splatting, a stochastic method for rendering massive 3DGS scenes. By sampling pixel-sized opaque points, splatting them atomically, and applying stochastic transparency, we eliminate the need for sorting. Our approach distributes workload evenly across GPU threads, enabling the real-time display of hundreds of millions of Gaussians efficiently.
Technical Paper
Research & Education
Generative AI
Geometry
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Computer Animation Festival
Full Conference Supporter
Full Conference
Description14-year-old Zain, a Palestinian swimming champion, faces war and man-made famine. He becomes the sole caregiver to his little brother. When aid drops into the sea, desperate for food, Zain is forced to swim—not for passion, but for survival.
Technical Paper
Research & Education
Geometry
Hardware
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionWe present a highly parallel GPU algorithm for 2D constrained Delaunay triangulation that robustly handles arbitrary valid constraints. It computes triangulations of up to 10 million vertices in 0.1 seconds on an RTX 4090, substantially outperforming prior GPU methods and widely used CPU implementations across diverse benchmark inputs.
Technical Paper
Research & Education
Animation
Fabrication
Haptics
Simulation
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionOptimal imaging performance requires designing optics with the downstream processing in the loop—especially as AI pipelines grow in importance. Yet the scalar nature of loss functions in the processing-aware setting breaks industry-standard lens design solvers. We generalize classical ray aberrations to task-driven objectives, bringing robust solvers to end-to-end design.
Technical Paper
Research & Education
Games
Real-Time
Rendering
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionWe introduce the Generalized Spherical Harmonics Product, a novel method supporting inputs with different SH orders and flexible output truncation. Using Spherical Grids as an intermediate representation, we reduce SH products to simple point-wise multiplications, achieving 3.5–6.5× speedup over state-of-the-art methods while maintaining accuracy for real-time rendering applications.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Rendering
Robotics
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Rendering
Robotics
Full Conference Supporter
Full Conference
DescriptionDeG is an image-to-3D generative framework for Gaussian splatting that learns to place detail adaptively. It evaluates each Gaussian’s effect on rendering loss, then reinforces density in important regions and suppresses less useful ones, enabling variable-resolution, high-quality 3D asset generation.
Technical Workshop
Research & Education
Generative AI
Full Conference Supporter
Full Conference
DescriptionGenerative AI is rapidly becoming a unifying interface for computer graphics, connecting large language models (LLMs), visual language models (VLMs), and video foundation models with geometry- and physics-aware representations. Led by experts from the University of Pennsylvania, MIT CSAIL, University of Utah, and UCLA, this workshop explores how modern generative models are reshaping the classical graphics pipeline, spanning 3D content creation, physics simulation, and animation. The program features a distinguished lineup of confirmed speakers from Stanford, the University of Chicago, Anuttacon, and Meshy AI, representing both academic pioneers and industry leaders.
The primary objective is to address the fundamental challenges of deploying generative methods in real-world production, specifically focusing on the controllability and editability of generated content, physical plausibility and consistency over time, and practical integration into professional workflows for film, games, and CAD. Attendees can expect a highly interactive session structure that moves beyond theory into practical application. The workshop includes five 30-minute invited talks and two 15-minute lightning talks highlighting emerging directions like garment synthesis and character animation. Most importantly, the session concludes with an 85-minute moderated panel and open audience discussion designed to synthesize themes, identify actionable open problems, and foster direct engagement between researchers and practitioners.
The primary objective is to address the fundamental challenges of deploying generative methods in real-world production, specifically focusing on the controllability and editability of generated content, physical plausibility and consistency over time, and practical integration into professional workflows for film, games, and CAD. Attendees can expect a highly interactive session structure that moves beyond theory into practical application. The workshop includes five 30-minute invited talks and two 15-minute lightning talks highlighting emerging directions like garment synthesis and character animation. Most importantly, the session concludes with an 85-minute moderated panel and open audience discussion designed to synthesize themes, identify actionable open problems, and foster direct engagement between researchers and practitioners.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Full Conference Supporter
Full Conference
DescriptionOrbit-Space Geometric Probability Paths (OGPP) is a flow-matching framework for generative modeling of particle systems that exploits shared geometric structure and uses probability path geometry to encode geometric information. It enables high-quality, efficient generation of particle configurations and point-cloud surfaces with normals, achieving state-of-the-art performance across diverse generation tasks.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Real-Time
Rendering
Scientific Visualization
Simulation
Full Conference Supporter
Full Conference
DescriptionWe introduce a highly efficient variational framework for computing optimal vortex cores in 3D unsteady flows. By unifying diverse detection criteria through time-preintegrated Lagrangians, our method accurately reconstructs spacetime vortex manifolds by solving Euler-Lagrange equations in just one time step, enabling rapid and robust flow analysis.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Full Conference Supporter
Full Conference
DescriptionWe present a physically grounded generative imaging system to resolve high-dimensional, time-resolved plenoptic light transport from sparse measurements. Using decoupled laser-detector capture, 3D foundation-model priors, and differentiable transient rendering, we reveals indirect light transport for applications including multi-bounce separation, time unwarping, and time-resolved relighting.
Technical Paper
Research & Education
Geometry
Modeling
Real-Time
Rendering
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Technical Paper
Research & Education
Generative AI
Geometry
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionGeoQuery is a geometry-guided single-step diffusion framework for mitigating query contamination in existing multi-view self attention mechanism. By retrieving proxy features from clean reference views via geometric correspondences, it bypasses corrupted query features and enables high-fidelity restoration for sparse-view 3D Gaussian Splatting.
Art Paper
Arts & Design
Art
Artificial Intelligence/Machine Learning
Audio
Augmented Reality
Digital Twins
Games
Geometry
Hardware
Performance
Real-Time
Scientific Visualization
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionThis work presents Gesture Flux, an immersive performance system integrating mixed-reality gesture recognition, Budaixi (traditional Taiwanese glove puppetry), and contemporary dance. Through three interaction modes, hidden puppeteering gestures are transformed from direct articulation to perceptual extension and energetic radiation, visualized as dynamic traces and extended into full-body performance, proposing a cross-disciplinary approach to cultural reinterpretation and embodied interaction.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Fabrication
Generative AI
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionGimmBO helps users interactively explore combinations of customized image generation styles to create desired results. It replaces manual slider tuning with a tailored Preferential Bayesian Optimization approach, outperforming alternatives in simulations and user studies. It further extends to combining content and scaling to large model collections by integrating model retrieval.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Fabrication
Geometry
Modeling
Full Conference Supporter
Full Conference
DescriptionWe propose a global optimization framework for simplifying quad-dominant meshes used in game and animation production. By formulating edge selection as an Integer Linear Programming problem with partial poly-chord segmentation and curvature-aware constraints, our method preserves clean edge flows and quad topology far better than existing greedy approaches.
Spatial Storytelling
Gaming & Interactive
New Technologies
Performance
Real-Time
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionThis talk chronicles the engineering and design journey of bringing Google Maps Immersive View to standalone XR. We detail the unique challenges of rendering this fully navigable, global-scale environment on constrained headsets. Highlighting rendering optimizations for 3D Tiles and 3D Gaussian Splatting, intuitive hands-only gestural navigation, and adaptive spatial audio, we reveal how to achieve performant photorealism and active user presence on next-generation XR hardware.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Fabrication
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionGMT is a Geometric Multigrid Transformer that significantly accelerates microstructure homogenization. Embedding Transformers into the multigrid hierarchy to capture multi-scale physical interactions delivers a 160x speedup over SOTA GPU solvers while maintaining engineering-grade accuracy at high resolutions. GMT enables real-time material screening, multi-scale simulations, and rapid inverse design of metamaterials.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Generative AI
Full Conference Supporter
Full Conference
DescriptionWe present Go-with-the-Track, a video generation framework conditioned on multiple reference images and point-tracks that jointly anchor the generated frames and the references. Our approach enables precise control over motion and multi-reference compositing, unlocking a wide range of applications, including video restylization, keypoint- and mesh-driven compositing, and camera motion control.
Immersive Pavilion
Arts & Design
Gaming & Interactive
New Technologies
Research & Education
Animation
Art
Artificial Intelligence/Machine Learning
Audio
Augmented Reality
Computer Vision
Digital Twins
Education
Ethics and Society
Fabrication
Games
Generative AI
Haptics
Hardware
Image Processing
Modeling
Performance
Pipeline Tools and Work
Real-Time
Scientific Visualization
Simulation
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionGoo Goo Clash is a high-energy "VE-Sport" that reimagines the classic "Snake" arcade game for Mixed Reality. Multiple players physically sprint within a shared arena using standalone headsets. By blending real-world athleticism with digital strategy, it transforms isolating VR into a thrilling, scalable, and co-located competitive spectator sport.
Art Paper
Arts & Design
Animation
Art
Artificial Intelligence/Machine Learning
Audio
Capture/Scanning
Computer Vision
Display
Ethics and Society
Generative AI
Hardware
Image Processing
Physical AI
Real-Time
Robotics
Scientific Visualization
Simulation
Spatial Computing
Full Conference Supporter
Full Conference
Experience
DescriptionGorgon Loop makes algorithmic judgement in public space visible as a collective, embodied process. Through machine vision, language models, and group commentary, it exposes how limited data, bias, and automation shape social evaluation, offering a critical lens on the cultural consequences of AI perception.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Robotics
Simulation
Full Conference Supporter
Full Conference
DescriptionWe present Generative Pretrained Controllers, a framework for training reusable generative controllers for physically simulated characters. GPC leverages FSQ discretization with end-to-end RL to learn reusable motor skills from large-scale motion data. Once trained, the generative controller can be finetuned and adapted to perform diverse downstream tasks using naturalistic behaviors.
Course
New Technologies
Rendering
Full Conference Supporter
Full Conference
DescriptionThis course teaches attendees how to use GPU Ray Tracing through NVIDIA’s OptiX API and the NVIDIA OptiX Wrappers Library (OWL). Attendees taking this course should be able to write their own GPU ray tracers, even without having to become experts in PTX/IR embedding or Shader Binding Tables.
Technical Paper
Research & Education
Geometry
Hardware
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionWe present a GPU-accelerated certified algorithm for the directed Hausdorff distance between triangle meshes. By replacing recursive branch-and-bound with a parallel wavefront pipeline, it delivers millisecond-scale, tolerance-controlled results on models with millions of triangles, enabling interactive geometric queries while closely matching state-of-the-art CPU accuracy on large, challenging benchmark datasets.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Lighting
Modeling
Real-Time
Rendering
Full Conference Supporter
Full Conference
DescriptionGR3EN enables relighting 3D environments to user-specified lighting conditions. It provides full control over every light source in a 3D scene — enabling adding new light sources, changing the color of existing ones or turning them off entirely. Shadows, reflections, and interreflections all update realistically.
Art Gallery
Arts & Design
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Full Conference Supporter
Full Conference
Experience
DescriptionGradi is a suite of speculative wearable intrafaces that exposes the seams in AI’s promise of effortless interaction. Each device examines a dominant logic shaping AI culture: calibration, compression, prediction, and mediation. Calibration holds you in permanent beta. Compression lets a proxy speak through your body. Prediction keeps you within the guardrails of the sayable. Mediation polishes speech into identityless fluency. Through interactive subversive artifacts and videos of strange use cases, Gradi challenges tacit expectations with quiet perversion, trading frictionless and phantasmagoric integration for situated misalignments. A permanent rehearsal that keeps the seam visible and felt.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Fabrication
Generative AI
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionWe generate 2D low-poly meshes that abstract images using score distillation sampling. We focus on ensuring the validity of the output mesh, quantized palette colors and semantic similarity to the input. The resulting vector images are easy to edit and fabricated as mosaics, embroidery, crocheting, patchwork and stencils.
Technical Paper
Research & Education
Geometry
Modeling
Rendering
Simulation
Full Conference Supporter
Full Conference
DescriptionWe present a gradient-domain Monte Carlo framework for solving Poisson equations on complex domains. The method directly estimates solution differences between query points and reconstructs the final solution efficiently, reducing variance and improving convergence without introducing additional bias beyond the underlying WoS-based solver.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionThis paper introduces a first numerical optimization method for approximately generating non-uniform low-discrepancy point sets with an arbitrary target distribution. It is based on a new differentiable estimator of non-uniform L2-discrepancy for a given point set.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionGraphical X Splatting (GraphiXS) is a novel probabilistic framework for systematically modelling data uncertainty in 4D Gaussian Splatting. It handles various types of spatial and temporal sparsity, including missing frames and sparse viewpoints, and generalizes across diverse primitives. Extensive evaluations show that GraphiXS outperforms existing methods across a range of sparse data scenarios.
Technical Workshop
New Technologies
Research & Education
Animation
Artificial Intelligence/Machine Learning
Computer Vision
Digital Twins
Dynamics
Education
Fabrication
Generative AI
Geometry
Image Processing
Math Foundations and Theory
Modeling
Physical AI
Rendering
Scientific Visualization
Simulation
Spatial Computing
Full Conference Supporter
Full Conference
DescriptionComputer graphics is increasingly central to modern scientific discovery, providing not only visualization but also physically grounded modeling, simulation, inverse problem formulations, and interactive systems for hypothesis generation. Building on the Graphics4Science course at SIGGRAPH 2025, Graphics4Science 2026 adopts a workshop-style format with fewer broadcast talks, deeper technical presentations, and more discussion. The program features four keynotes spanning computational imaging, scientific reconstruction for structural biology, generative modeling for molecular structures, and reduced-order modeling for physical simulation, complemented by two curated lightning-talk sessions highlighting advances from the community.
As graphics-based methods move from retrospective analysis toward guiding experiments and engineering decisions, a shared requirement across these areas is reliability. In this workshop, reliable scientific instruments refers to methods and systems that remain useful under scientific constraints, including limited or noisy measurements, distribution shift across instruments and conditions, and the need to communicate uncertainty.
This notion of reliability includes physical and geometric validity, numerical stability under perturbations, calibrated uncertainty estimates, and clearly stated assumptions and failure modes. It also includes practical concerns such as reproducible pipelines, transparent reporting, and interfaces that allow scientists to interrogate and stress-test results rather than treating models as black boxes.
Against this backdrop, Graphics4Science 2026 aims to move beyond a catalog of applications toward an actionable technical agenda for graphics-enabled science. A unifying theme of the workshop is translation: connecting computational methods to scientific constraints, evaluating progress when ground truth is limited, and packaging techniques into practical tools that domain experts can trust, use, and iterate on.
As graphics-based methods move from retrospective analysis toward guiding experiments and engineering decisions, a shared requirement across these areas is reliability. In this workshop, reliable scientific instruments refers to methods and systems that remain useful under scientific constraints, including limited or noisy measurements, distribution shift across instruments and conditions, and the need to communicate uncertainty.
This notion of reliability includes physical and geometric validity, numerical stability under perturbations, calibrated uncertainty estimates, and clearly stated assumptions and failure modes. It also includes practical concerns such as reproducible pipelines, transparent reporting, and interfaces that allow scientists to interrogate and stress-test results rather than treating models as black boxes.
Against this backdrop, Graphics4Science 2026 aims to move beyond a catalog of applications toward an actionable technical agenda for graphics-enabled science. A unifying theme of the workshop is translation: connecting computational methods to scientific constraints, evaluating progress when ground truth is limited, and packaging techniques into practical tools that domain experts can trust, use, and iterate on.
Emerging Technologies
Arts & Design
Gaming & Interactive
New Technologies
Research & Education
Art
Artificial Intelligence/Machine Learning
Audio
Augmented Reality
Capture/Scanning
Computer Vision
Digital Twins
Display
Dynamics
Education
Fabrication
Games
Generative AI
Graphics Systems Architecture
Haptics
Hardware
Image Processing
Performance
Physical AI
Pipeline Tools and Work
Real-Time
Robotics
Scientific Visualization
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionThis study proposes a floor-mounted turntable that enhances impact sensation in VR by simulating reaction forces to the feet during striking actions. Unlike conventional hand-based haptic devices, the proposed approach leverages whole-body force propagation. We demonstrate a VR-based sports application that enhances impact realism through instantaneous rotational feedback.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Full Conference Supporter
Full Conference
DescriptionWe introduce and demonstrate the first closed-loop adaptive optics system capable of optically correcting aberrations in real-time without a guidestar or a wavefront sensor. We enable these capabilities by combining asymmetric apertures, machine learning–based PSF estimation and phase retrieval algorithms, and a high resolution spatial light modulator.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionHairGPT introduces a paradigm shift toward strand-as-language modeling, reformulating 3D realistic hair generation as a dual-decoupled autoregressive process where strands are treated as the fundamental generative units.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionHairLRM integrates Large Reconstruction Model (LRM) geometric priors into strand generation. Leveraging an LRM mesh anchor and a Dual Orientation AutoEncoder, it transforms coarse geometry into high-fidelity strands, disentangling complex topologies to set a new benchmark for robust, accurate hair reconstruction.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionHairPort is a 3D-aware hairstyle transfer method that faithfully moves a reference hairstyle onto a source face image, even under large pose and scale differences. By separating hair removal from synthesis and enforcing geometric consistency through 3D reconstruction, HairPort achieves accurate, identity-preserving results that outperform existing approaches.
Course
New Technologies
Artificial Intelligence/Machine Learning
Digital Twins
Physical AI
Pipeline Tools and Work
Real-Time
Robotics
Simulation
Full Conference Supporter
Full Conference
Experience
DescriptionHands-on course on Newton, a GPU-accelerated physics engine, and its integration with Isaac Lab for robot learning. Covers Newton’s architecture and import pipeline, solver and contact tuning, articulation and sensors, then training and evaluating policies in Isaac Lab with Newton as the physics backend.
Course
New Technologies
Research & Education
Capture/Scanning
Computer Vision
Digital Twins
Physical AI
Simulation
Full Conference Supporter
Full Conference
Experience
DescriptionLearn how to reconstruct a large scene for robotics testing using NVIDIA Omniverse NuRec Gaussian-based reconstruction technologies to perform multi-GPU training, object segmentation, and extraction. This lab will walk through core concepts with a step-by-step workflow for data capture, reconstruction, and object level-integration for robotics simulation in NVIDIA Isaac Lab.
Course
Gaming & Interactive
Production & Animation
Animation
Games
Performance
Pipeline Tools and Work
Full Conference Supporter
Full Conference
Experience
DescriptionThis hands-on course focuses on debugging USD composition in production. Participants gain practical experience authoring data across composition arcs and diagnosing issues in complex scenes. Building on USD in Production, the course emphasizes interactive use of modern debugging and visualization tools to understand how opinions, overrides, and data resolve.
Course
Arts & Design
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Artificial Intelligence/Machine Learning
Education
Generative AI
Rendering
Full Conference Supporter
Full Conference
Experience
DescriptionMove beyond static pixels into explorable, playable worlds, generative images, and generative video. Join Google DeepMind engineers as we get hands-on, using NanoBanana Pro for reasoned concept art, Veo 3.1 for video generation, and Genie 3 to turn images and text into playable, interactive worlds.
Course
Arts & Design
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Animation
Artificial Intelligence/Machine Learning
Capture/Scanning
Physical AI
Simulation
Full Conference Supporter
Full Conference
Experience
DescriptionWe introduce a novel capture-to-simulation pipeline for in-the-wild 3D Gaussian Splat scenes,
showing our research-based solutions for segmenting objects, predicting their volumetric mechanical properties using a recent learning-based method, and mixed splat-mesh physics simulations, all supported by recent features in the Kaolin Library.
showing our research-based solutions for segmenting objects, predicting their volumetric mechanical properties using a recent learning-based method, and mixed splat-mesh physics simulations, all supported by recent features in the Kaolin Library.
Course
New Technologies
Artificial Intelligence/Machine Learning
Computer Vision
Generative AI
Real-Time
Full Conference Supporter
Full Conference
Experience
DescriptionLearn to build GPU ML inference pipelines with Slang. Use a lightweight, open set of pre-written kernels to implement an MNIST CNN from scratch: load SafeTensors weights, run inference, and apply kernel fusion for better performance. Hands-on, cross-platform, and highly customizable—ideal for games, tools, and research.
Course
New Technologies
Artificial Intelligence/Machine Learning
Computer Vision
Education
Image Processing
Math Foundations and Theory
Performance
Simulation
Full Conference Supporter
Full Conference
Experience
DescriptionIn this hands-on programming course you'll learn how to write and run quantum computing code. Then you'll program the quantum teleportation algorithm, and run it on a high-quality simulator to transfer the state of a quantum bit to your partner, who will decode it to get your secret message.
Course
Gaming & Interactive
New Technologies
Artificial Intelligence/Machine Learning
Games
Graphics Systems Architecture
Performance
Real-Time
Rendering
Full Conference Supporter
Full Conference
Experience
DescriptionThis hands-on course introduces Slang, an open-source, open governance shading language hosted by Khronos that simplifies graphics development across platforms. Designed to tackle the growing complexity of shader code, Slang offers modern programming constructs while maintaining top performance on current GPUs.
Course
New Technologies
Research & Education
Digital Twins
Physical AI
Robotics
Simulation
Full Conference Supporter
Full Conference
Experience
DescriptionHands-on course on preparing robot USD assets for simulation, covering why physics tuning is foundational to robot learning, inspecting and configuring colliders and joints with OpenUSD/PhysX, and tuning control gains for stable, realistic motion. Participants work with a robot hand asset and apply best practices for simulation-ready authoring.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Augmented Reality
Generative AI
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionHeadRouter is a training-free image editing framework tailored for Multimodal Diffusion Transformers (MM-DiTs). It addresses diminishing text guidance by adaptively routing attention heads based on semantic sensitivity and refining token representations. The method improves semantic alignment and structural preservation over state-of-the-art baselines through dynamic head-level control without additional training mechanisms.
Computer Animation Festival
Full Conference Supporter
Full Conference
Descriptionn 2007, former World War II Heart Mountain Incarceration Camp incarceree Shig Yabu and former Topaz Incarceration Camp incarceree Willie Ito published the children's picture book, Hello Maggie! Yabu wrote the autobiographical story about how he was forced to give away his pets when he was removed from his home in San Francisco, and how he eventually adopted a baby magpie in Wyoming. The bird, which he named Maggie, became Yabu’s constant companion during his time in camp, and eventually learned to mimic human speech. Ito, a veteran artist and animator at Disney, Warner Brothers, and Hanna-Barbera, illustrated the story.
In 2025 the Heart Mountain Wyoming Foundation produced a 23 minute animated version of the classic children's book.
In 2025 the Heart Mountain Wyoming Foundation produced a 23 minute animated version of the classic children's book.
Technical Paper
Research & Education
Animation
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionHeterogeneous Subspace Corrections (HSC) accelerates GPU simulations of complex multibody systems. By decoupling deformable and rigid/affine components into separate Neumann-preconditioned Krylov iterations and utilizing an Adaptive Cross Approximation (ACA) for low-rank coupling, it overcomes ill-conditioning to achieve direct-solver convergence speeds for stiff, high-resolution models.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionThis work develops a hyperspectral imager using a SPAD image sensor that achieves a spatial resolution of 1024 × 768 pixels across 123 spectral bands spanning visible and near-infrared wavelengths, at a temporal rate of 26 frames per second.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Geometry
Modeling
Rendering
Simulation
Full Conference Supporter
Full Conference
DescriptionWe present a 4D (spatio-temporal) cloth capture system and a two-level pattern that achieves 1mm spatial resolution using only 16 RGB cameras. Our method produces temporally coherent sequences with no self penetration that faithfully capture fine wrinkles and folds even during complex motions with self-contact.
Technical Paper
Research & Education
Animation
Geometry
Hardware
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionWe propose a conservative algorithm to test the geometrical validity of polynomial finite elements as they deform linearly in time. In elastodynamic simulation, our algorithm guarantees that the system remains physically valid during the entire trajectory, not only at discrete time steps, even when implemented using floating point arithmetic.
ACM SIGGRAPH 365
Arts & Design
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Animation
Art
Artificial Intelligence/Machine Learning
Audio
Augmented Reality
Capture/Scanning
Computer Vision
Digital Twins
Display
Dynamics
Education
Ethics and Society
Fabrication
Games
Generative AI
Geometry
Graphics Systems Architecture
Haptics
Hardware
Image Processing
Industry Insight
Lighting
Math Foundations and Theory
Modeling
Performance
Physical AI
Pipeline Tools and Work
Real-Time
Rendering
Robotics
Scientific Visualization
Simulation
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionIf you are interested in the history of computer graphics, this session is for you. The ACM SIGGRAPH History Archive is both a physical collection and an online archive of SIGGRAPH organization leadership and events, conference presentations, demonstrations, animation screenings, art exhibitions, and more. The archive is a volunteer effort involving pioneers, professionals, students, and community members. This session will showcase the new additions to the archive and include short talks by student interns and volunteers. We will also focus on how this incredibly information-rich archive has fueled the future of computer graphics research and creative endeavors.
The archive has grown tremendously and now includes more than 75,000 pages of information. One of the most significant contributions to the community is opening up the archive so the AI bots could learn from 52 years worth of conference presentations, demonstrations, animation screening, etc. Today, Gemini, ChatGPT and other AI assistants often cite the ACM SIGGRAPH History Archive and the information related to the history of computer graphics is much more accurate. Another significant addition was the programming of an online inventory system that tracks the 4000+ physical artifacts in the collection. Eighteen years worth of SIGGRAPH Asia conference content was also added to the archive along with the unique programs from SIGGRAPH's past such as performances, competitions, and special sessions and events. Overall, this session is an exciting way to capture a glimpse into SIGGRAPH's past and speculate on how access to this information will alter computer graphics' future trajectory.
The archive has grown tremendously and now includes more than 75,000 pages of information. One of the most significant contributions to the community is opening up the archive so the AI bots could learn from 52 years worth of conference presentations, demonstrations, animation screening, etc. Today, Gemini, ChatGPT and other AI assistants often cite the ACM SIGGRAPH History Archive and the information related to the history of computer graphics is much more accurate. Another significant addition was the programming of an online inventory system that tracks the 4000+ physical artifacts in the collection. Eighteen years worth of SIGGRAPH Asia conference content was also added to the archive along with the unique programs from SIGGRAPH's past such as performances, competitions, and special sessions and events. Overall, this session is an exciting way to capture a glimpse into SIGGRAPH's past and speculate on how access to this information will alter computer graphics' future trajectory.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Generative AI
Rendering
Full Conference Supporter
Full Conference
DescriptionWe introduce HL-OutPaint, a coarse-to-fine framework for high-resolution long-range video outpainting. HL-OutPaint incorporates a novel global-local frame swapping mechanism that enables both long-term and short-term temporal consistency. Experimental results show that our approach produces spatially and temporally coherent results across diverse real-world high-resolution long-range videos.
Technical Paper
Research & Education
Augmented Reality
Display
Fabrication
Hardware
Real-Time
Rendering
Simulation
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Technical Paper
Research & Education
Augmented Reality
Display
Fabrication
Hardware
Real-Time
Rendering
Simulation
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionTo bring physically accurate 3D cues to holography, HoloPathTracer moves beyond prior computer-generated holography pipelines that separate radiance rendering from wave propagation. By tracing complex wave paths directly through the scene, it enables phase-only holograms with natural defocus, reflections, refractions, and more faithful visual realism.
Course
New Technologies
Research & Education
Artificial Intelligence/Machine Learning
Computer Vision
Digital Twins
Hardware
Physical AI
Pipeline Tools and Work
Real-Time
Robotics
Simulation
Full Conference Supporter
Full Conference
DescriptionThis course covers building end-to-end physical AI for humanoid and general robots. Covers data generation with NVIDIA Cosmos (Predict, Transfer) and synthetic data; training with Isaac Lab (RL/imitation) and GR00T VLAs; and deployment on the edge with Jetson Thor. Instructors: Edith Llontop, Akul Santhosh.
ACM SIGGRAPH 365
Arts & Design
Gaming & Interactive
Production & Animation
Research & Education
Animation
Art
Education
Games
Lighting
Modeling
Performance
Pipeline Tools and Work
Real-Time
Full Conference Supporter
Full Conference
Experience
DescriptionThe ACM SIGGRAPH Early Career Development Committee’s “Resume and Reel Review” program has long offered students and early career professionals the chance to have their work reviewed by industry experts at SIGGRAPH conferences.
This session will feature a panel of industry professionals who will discuss and review a selection from this year’s program live. This unique opportunity allows attendees to see real-time reviews, learn what makes a great resume and reel, and ask questions directly to experts.
Register for a one-on-one session at https://ecdc.siggraph.org.
This session will feature a panel of industry professionals who will discuss and review a selection from this year’s program live. This unique opportunity allows attendees to see real-time reviews, learn what makes a great resume and reel, and ask questions directly to experts.
Register for a one-on-one session at https://ecdc.siggraph.org.
Course
Gaming & Interactive
Education
Games
Real-Time
Rendering
Full Conference Supporter
Full Conference
DescriptionThis Khronos course covers major Vulkan improvements since version 1.0, focusing on extensions such as descriptor heaps and dynamic rendering that improve developer productivity. It introduces key Vulkan concepts and provides comprehensive guidance for writing and modernizing Vulkan applications in 2026, focusing on resource, pipeline, draw, and synchronization management.
Technical Workshop
Arts & Design
Research & Education
Art
Artificial Intelligence/Machine Learning
Generative AI
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionRecent advances in generative AI are transforming how visual and spatial content is created, enabling new forms of collaboration between humans and intelligent systems. This workshop explores human–AI co-creation in generative art, where AI acts not merely as an automation tool but as a creative partner supporting exploration, iteration, and artistic expression. The workshop brings together researchers, artists, and engineers from computer graphics, machine learning, human–computer interaction, and computational design to examine emerging systems for collaborative creativity. Topics include interactive creative tools, generative models for visual and 3D art, and fabrication-aware design systems that connect digital creation with real-world production. The workshop will feature keynote talks and interactive discussions exploring questions such as artistic control in generative systems, effective interfaces for co-creation, and integrating fabrication constraints into generative design, aiming to inspire new directions for human–AI collaboration in generative art and design. More details in: https://human-ai-cocreation.github.io/
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Generative AI
Modeling
Full Conference Supporter
Full Conference
DescriptionHumanFlow is a unified flow-matching framework for controllable full-body human image generation. It introduces a Control Encoder, Token-ControlNet, and topology-aware loss for structural consistency, along with the large-scale MiCoGen dataset, achieving improved controllability and fidelity over existing diffusion and flow-based methods.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Generative AI
Geometry
Modeling
Simulation
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Real-Time
Rendering
Full Conference Supporter
Full Conference
DescriptionWe present a multi-class Gaussian Splatting Wang Tiles framework for synthesizing large, detailed scenes from multiple exemplars. Our method enables smooth transitions via hybrid tile mixing, supports real-time rendering with hybrid CPU–GPU sorting, and provides an interactive authoring tool for editing class distributions and scene structure.
Course
Production & Animation
Animation
Industry Insight
Pipeline Tools and Work
Real-Time
Rendering
Full Conference Supporter
Full Conference
DescriptionBased on real production examples, Hydra in Production will present best practices and case studies of how to integrate OpenUSD's Hydra library into a content creation tool or pipeline, and leverage it to solve complex pipeline challenges and enable powerful workflows. This is a companion course to USD in Production.
Talks
Production & Animation
Animation
Artificial Intelligence/Machine Learning
Computer Vision
Graphics Systems Architecture
Image Processing
Lighting
Performance
Pipeline Tools and Work
Rendering
Full Conference Supporter
Full Conference
DescriptionHydraOps is the first production, artist-facing lighting framework built on Hydra 2.0 within OpenDCC, unifying interactive and final rendering from a single USD scene. Replacing a legacy in-house engine with Hydra’s modular scene index architecture, it streamlines evaluation, reduces complexity, and lowers maintenance costs.
Talks
Production & Animation
Research & Education
Animation
Artificial Intelligence/Machine Learning
Audio
Computer Vision
Performance
Full Conference Supporter
Full Conference
DescriptionThis talk explores how Wētā FX’s Anatomically Plausible Facial System enabled Varang’s ferocious, emotionally authentic performance in Avatar: Fire and Ash. Using Neytiri as a case study, we examine the motivations behind a revolutionary shift in facial technology and end-to-end development challenges required for Varang to stand alongside legacy characters.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Augmented Reality
Generative AI
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Generative AI
Modeling
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Technical Paper
Research & Education
Display
Generative AI
Hardware
Image Processing
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Full Conference Supporter
Full Conference
DescriptionImg2CADSeq is a multi-stage pipeline converting single-view images into parametrically editable 3D CAD models. Using a coarse-to-fine hierarchical codebook and VQ-Diffusion , it bridges 2D visual features with 3D semantics , generating standard STEP files ready for direct commercial software use.
Technical Paper
Research & Education
Generative AI
Geometry
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionWe propose a 3D Gaussian Splatting reconstruction method with immediate feedback that handles unordered image captures and very large scenes. Through fast place-recognition-driven matching, cluster-based loop closure with graph-propagated correction, and a progressive Gaussian hierarchy construction, it achieves robust, scalable reconstruction at a fraction of offline processing times.
Immersive Pavilion
Arts & Design
Gaming & Interactive
New Technologies
Research & Education
Animation
Art
Artificial Intelligence/Machine Learning
Audio
Augmented Reality
Computer Vision
Digital Twins
Education
Ethics and Society
Fabrication
Games
Generative AI
Haptics
Hardware
Image Processing
Modeling
Performance
Pipeline Tools and Work
Real-Time
Scientific Visualization
Simulation
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionThis VR-based instructional program trains learners in mouse cheek blood collection under the 3Rs principles. It bridges classroom instruction and laboratory practice by allowing novices to rehearse procedures safely in a virtual environment, while the simulation familiarizes them with settings prior to on-site training, enhancing readiness and reducing animal distress.
Technical Paper
Research & Education
Geometry
Modeling
Full Conference Supporter
Full Conference
DescriptionWe introduce an implicit representation of continuous, bijective, orientation-preserving maps between genus zero surfaces with or without boundary. The distortion of these maps can easily be minimized by optimizing the Ginzburg-Landau functional---a ubiquitous model in physics and differential geometry---leading to a simple algorithm for computing bijective correspondences using only standard tools of the tangent vector field toolbox. The method avoids combinatorial mesh modifications and does not require barrier functions to enforce bijectivity making it more robust to noise and simpler to implement.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Geometry
Modeling
Rendering
Simulation
Full Conference Supporter
Full Conference
DescriptionWe present a real-time 4D compression framework for volumetric capture using the Truncated Signed Distance Field (TSDF) as coding domain. Combining DCT-based spatial coding with motion-compensated temporal prediction, our training-free GPU-accelerated system enables real-time encoding and decoding while substantially reducing bitrate for dynamic 3D content.
Immersive Pavilion
Arts & Design
Gaming & Interactive
New Technologies
Research & Education
Animation
Art
Artificial Intelligence/Machine Learning
Audio
Augmented Reality
Computer Vision
Digital Twins
Education
Ethics and Society
Fabrication
Games
Generative AI
Haptics
Hardware
Image Processing
Modeling
Performance
Pipeline Tools and Work
Real-Time
Scientific Visualization
Simulation
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionThis immersive installation invites attendees to directly engage with Impulse, an artist-made Unreal Engine tool for real-time choreography. Through hands-on motion capture and mixed reality visualization, participants can record movement, apply it across multiple digital dancers, and experiment with spatial composition in real time.
Talks
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Animation
Artificial Intelligence/Machine Learning
Computer Vision
Education
Games
Generative AI
Image Processing
Simulation
Full Conference Supporter
Full Conference
DescriptionWhen teaching 3D rotations, the prevailing rationale is to avoid Euler angles due to gimbal lock. In many game situations, quaternion slerp produces undesirable effects while Euler angle interpolation actually results in the correct behavior. Students should learn to select rotation representations that produce appropriate motion for the application.
Art Gallery
Arts & Design
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Full Conference Supporter
Full Conference
Experience
Description“in-between" is a modular installation about the immigrant experience and the difficulty of defining home. Born in Philippines and raised across China, Japan, and United States, I translate my own migration history—and seven interviews with Asian diaspora who moved to Tokyo—into fragile material and participatory data portraiture. The work unfolds in three parts: “deportation”, a two-hour performance where water drips onto washi printed with my life-journey at a cadence driven by worldwide searches for “deportation”; “home:1–8”, silkscreened interview portraits mounted in space; and “home:collective”, where visitors add their journeys and are paired with a stranger whose definition of home resonates.
Art Paper
Arts & Design
Research & Education
Animation
Art
Augmented Reality
Dynamics
Hardware
Lighting
Performance
Real-Time
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionThis work interweaves vision science with perceptual experience in art. Leveraging MR technology, it renders the infant's mysterious vision into a first-person experiential artwork. For the SIGGRAPH community, the project serves as a case study of how scientific theories and domain knowledge can be transformed into an embodied, explorable experience. Through exhibition, we found that Infant Vision opens other senses and encourages participants to reimagine the world.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Generative AI
Geometry
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionWe introduce InfiniteDiffusion, a training-free algorithm enabling diffusion models to generate spatially infinite outputs with the core properties of procedural noise: seed-consistency (determinism) and constant-time random access. We incorporate a novel architecture for terrain generation to produce Terrain Diffusion, a learned, real-time, infinite terrain generator with a procedural noise-like interface.
ACM SIGGRAPH 365
Arts & Design
Research & Education
Art
Education
Full Conference Supporter
Full Conference
Experience
DescriptionLooking for ways to connect with SIGGRAPH’s creative side? This session offers an open and inviting introduction to the ACM SIGGRAPH Digital Arts Community (DAC), a vibrant network that brings together artists, designers, technologists, and researchers working across digital and computational media arts.
Attendees will get a clear overview of how DAC supports creative exchange through inclusive programs designed to encourage experimentation, dialogue, and collaboration across disciplines. The session highlights key DAC initiatives, including SPARKS Lightning Talks, online and hybrid exhibitions, and the student digital art competition, along with opportunities for global engagement through partnerships with organizations such as ISEA and Expanded Animation.
Whether you are a student, a first-time SIGGRAPH attendee, or someone curious about engaging more deeply with SIGGRAPH’s creative communities, this session provides a friendly entry point into DAC’s activities and a chance to discover how you can share your work, connect with others, and participate in an international network of digital arts practitioners.
Attendees will get a clear overview of how DAC supports creative exchange through inclusive programs designed to encourage experimentation, dialogue, and collaboration across disciplines. The session highlights key DAC initiatives, including SPARKS Lightning Talks, online and hybrid exhibitions, and the student digital art competition, along with opportunities for global engagement through partnerships with organizations such as ISEA and Expanded Animation.
Whether you are a student, a first-time SIGGRAPH attendee, or someone curious about engaging more deeply with SIGGRAPH’s creative communities, this session provides a friendly entry point into DAC’s activities and a chance to discover how you can share your work, connect with others, and participate in an international network of digital arts practitioners.
Spatial Storytelling
Arts & Design
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Animation
Art
Education
Performance
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionA live Spatial Storytelling performance that pairs a short VR scene with a real-time “backstage reveal.” One performer acts in VR while a company member projects the live experience onto the screen, then demonstrates the hidden tools that drive the show, including triggers, actor menus, and control systems.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Generative AI
Modeling
Full Conference Supporter
Full Conference
DescriptionInspiration Seeds is a generative framework that combines two input images into diverse, visually coherent compositions—without text prompts. Trained on synthetic triplets using CLIP Sparse Autoencoders, it uncovers latent visual relationships between inputs, supporting early-stage creative exploration where designers seek emergent ideas from loosely connected visual references.
Talks
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Animation
Dynamics
Games
Generative AI
Geometry
Modeling
Performance
Pipeline Tools and Work
Real-Time
Simulation
Full Conference Supporter
Full Conference
DescriptionPlants in recent films at DreamWorks Animation are complex and stylized and may not conform to hierarchical and physical structures. From just input models of plants, our environment rigging toolset can produce rigs procedurally and interactively allowing the artist to control the process and account for the stylization and complexity.
Technical Paper
Research & Education
Animation
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionWe propose a novel framework utilizing a generalized Douglas-Rachford Splitting (DRS) scheme and hierarchical decomposition for high-resolution knitwear simulation. By convexifying complex yarn dynamics, it ensures robust, optimal convergence. With matrix-free GPU parallelization, our method achieves interactive simulation rates for multi-million DOF garments.
Technical Paper
Research & Education
Lighting
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Technical Paper
Research & Education
Lighting
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionWe propose a gradient-based discrete tomography method that models each voxel as a probability distribution over known materials, minimizing an object-space loss for projection consistency. Inspired by inverse rendering, it outperforms classical methods, supports scattering, and excels in sparse and limited-angle scenarios.
Technical Paper
Research & Education
Augmented Reality
Display
Fabrication
Hardware
Real-Time
Rendering
Simulation
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionWe demonstrate an invisible holographic window: a transparent surface‑relief computer-generated hologram formed on glass. Using a scattering‑suppressed encoding scheme and spatial‑frequency band‑division multiplexing, we achieve high‑transmittance, crosstalk‑free full‑color 3D image reconstruction, allowing clear viewing of a photorealistic holographic image overlaid in front of real‑world scenes through the glass.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Full Conference Supporter
Full Conference
DescriptionWe propose InvSculpt, which decomposes a mesh into a base shape and reusable details represented by a vector displacement field. It uses text-guided 2D priors and a 3D flow model for mask-free detail removal while preserving identity. The method enables accurate decomposition and high-fidelity detail transfer.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Full Conference Supporter
Full Conference
DescriptionIris3D is a novel 3D generation system that combines feedforward reconstruction models with a new Synchronized Diffusion Distillation (SDD) method to produce detailed geometry and vivid textures. By addressing view inconsistency and over-smoothing issues in traditional Score Distillation Sampling (SDS) methods, it enables controllable and high-quality 3D content creation.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Full Conference Supporter
Full Conference
DescriptionIskra is a system for differentiating through solutions to geometry processing problems.
It differentiates a broad class of algorithms, exploiting existing fast problem-specific schemes common to geometry processing, including local-global and ADMM solvers. It is compatible with machine learning frameworks, opening doors to new classes of inverse geometry processing applications.
It differentiates a broad class of algorithms, exploiting existing fast problem-specific schemes common to geometry processing, including local-global and ADMM solvers. It is compatible with machine learning frameworks, opening doors to new classes of inverse geometry processing applications.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Fabrication
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionWe present a computational method to design rigid-deployable Kirigami sheets based on isohedral tilings. By exploring combined topology and geometry spaces, our approach identifies self-deployable structures with controlled expansion. Our approach generates diverse designs, demonstrated through an interactive browser and validated with 3D-printed prototypes.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Augmented Reality
Rendering
Simulation
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionWe present a method for enhancing the perceived quality of foveated images by extrapolating local image statistics, such as intensity, orientation and especially phase, from coarse to fine scales. In particular, we demonstrate that correct phase alignment across space and scales is crucial for achieving perceptually convincing image quality.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Fabrication
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionJGS2-GQ replaces data-driven Cubature with training-free Gaussian Quadrature, enabling stable, high-resolution GPU simulation. It handles frictional contact, maintains Newton-like convergence, and outperforms JGS2 by 50% on novel deformations without training.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Games
Generative AI
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionJUST-DUB-IT is a unified audio-video diffusion model for automatic video dubbing. It jointly synthesizes translated speech and synchronized lip movements via a lightweight LoRA adapter, replacing traditional pipeline approaches. Trained on synthetically generated multilingual data with strategic inpainting, it achieves superior visual quality and lip sync across varied motion scenarios.
Emerging Technologies
Arts & Design
Gaming & Interactive
New Technologies
Research & Education
Art
Artificial Intelligence/Machine Learning
Audio
Augmented Reality
Capture/Scanning
Computer Vision
Digital Twins
Display
Dynamics
Education
Fabrication
Games
Generative AI
Graphics Systems Architecture
Haptics
Hardware
Image Processing
Performance
Physical AI
Pipeline Tools and Work
Real-Time
Robotics
Scientific Visualization
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionThis paper presents Katakko, a modular robot system that enables users to create and embody personalized social agents. Katakko provides communication-oriented modules (display, actuator, mobile, and sensor) and an optimization-based algorithm that automatically generates joint-to-joint mappings between the user's body and an arbitrarily assembled robot, with incremental updates upon reconfiguration.
Technical Paper
Research & Education
Fabrication
Geometry
Modeling
Robotics
Simulation
Full Conference Supporter
Full Conference
DescriptionKinematic Kitbashing synthesizes articulated 3D objects by assembling reusable parts according to an abstract kinematic graph. It preserves plausible attachments across motion using kinematics-aware geometric cues and can optimize for functional goals, enabling new articulated designs from existing part libraries.
Talks
Arts & Design
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Art
Artificial Intelligence/Machine Learning
Audio
Computer Vision
Digital Twins
Education
Fabrication
Games
Generative AI
Geometry
Industry Insight
Modeling
Performance
Pipeline Tools and Work
Real-Time
Rendering
Full Conference Supporter
Full Conference
DescriptionWhat if anyone who can draw could knit? This talk follows VARIANT3D’s journey building software, languages, and novel production systems that make 3D knitting a more democratic medium—enabling designers to create and manufacture curved, complex surfaces with unprecedented accessibility and control.
Spatial Storytelling
Spatial Computing
Full Conference Supporter
Full Conference
Experience
DescriptionLA Folded: A Cybernetic Mirage is an interactive, generative art installation that re-imagines Los Angeles as a living, ever-evolving digital twin. Conceived for SIGGRAPH 2026, the project bridges LA’s rich cinematic legacy with the frontier of spatial computing. By seamlessly integrating Kinect motion tracking and AI-generated point clouds, it invites participants to co-author a collaborative digital monument. As users input their personal memories and desires of LA, individual landmarks dynamically merge, overlap, and fold into a singular, massive 3D cyber environment—blurring the boundaries between physical presence, collective memory, and cybernetic space.
Credit：
Visual & Interactive Director: Jimmz ZHANG, Digital FUN & TEA Community
Technical Director: Ruokun CHEN, TEA Community & FLARE Lab
Sound Artist: Louzhang, USC School of Cinematic Arts
Credit：
Visual & Interactive Director: Jimmz ZHANG, Digital FUN & TEA Community
Technical Director: Ruokun CHEN, TEA Community & FLARE Lab
Sound Artist: Louzhang, USC School of Cinematic Arts
Spatial Storytelling
Art
Spatial Computing
Full Conference Supporter
Full Conference
Experience
DescriptionThe Spatial Storytelling LED Wall Showcase features a curated collection of short-form digital works by Los Angeles–based artists. Spanning immersive media, experimental film, generative art, and digital storytelling, the showcase highlights creative practices that explore new relationships between narrative, technology, and space.
Join us between sessions and during the dedicated lunch hour to experience these works at architectural scale and celebrate the vibrant artistic community of Los Angeles within the Spatial Storytelling program.
LA Local Artists:
Narae Kim
Artwork Title: Flowly
Learn more: https://www.kimnarae.com
Yara Feghali
Artwork Title: Be.Longing XR
Michaela Ternasky-Holland
Artwork Title: Kapwa
Learn more:
@michaelaternaskyholland / www.michaelaternaskyholland.com
@sisiigodly / https://aaron.work/
Faith “Aya” Umoh
Artwork Title: Noösphere
Learn more: www.FaithAyaUmoh.com
Isabel Beavers
Artwork Title: Aphrodite's Dream
Learn more: www.isabelbeavers.com / @isabelbeavers
Wadooah Wali
Artwork Title: The Obscura Society
Learn more: www.newcanvas.co / www.instagram.com/newcanvas_xr / www.linkedin.com/company/newcanvas
Huntrezz Janos
Artwork Title: AfroHungaria
Learn more: www.huntrezz.com
Scarlett Kim
Artwork Title: BEASTS
Learn more: scarlett.kim
Eugene Yen
Artwork Title: The Evolution of My Rectangles
Learn more: www.eugeneyen.com
Join us between sessions and during the dedicated lunch hour to experience these works at architectural scale and celebrate the vibrant artistic community of Los Angeles within the Spatial Storytelling program.
LA Local Artists:
Narae Kim
Artwork Title: Flowly
Learn more: https://www.kimnarae.com
Yara Feghali
Artwork Title: Be.Longing XR
Michaela Ternasky-Holland
Artwork Title: Kapwa
Learn more:
@michaelaternaskyholland / www.michaelaternaskyholland.com
@sisiigodly / https://aaron.work/
Faith “Aya” Umoh
Artwork Title: Noösphere
Learn more: www.FaithAyaUmoh.com
Isabel Beavers
Artwork Title: Aphrodite's Dream
Learn more: www.isabelbeavers.com / @isabelbeavers
Wadooah Wali
Artwork Title: The Obscura Society
Learn more: www.newcanvas.co / www.instagram.com/newcanvas_xr / www.linkedin.com/company/newcanvas
Huntrezz Janos
Artwork Title: AfroHungaria
Learn more: www.huntrezz.com
Scarlett Kim
Artwork Title: BEASTS
Learn more: scarlett.kim
Eugene Yen
Artwork Title: The Evolution of My Rectangles
Learn more: www.eugeneyen.com
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Real-Time
Rendering
Scientific Visualization
Simulation
Full Conference Supporter
Full Conference
DescriptionLagrangianSplats reconstructs physically plausible 3D fluid velocity fields from sparse video by combining divergence-free kernels with Lagrangian Gaussian transport. A sliding-window optimization enables efficient long-range supervision, producing state-of-the-art transport consistency and physical accuracy on synthetic and real fluid datasets.
Immersive Pavilion
Arts & Design
Gaming & Interactive
New Technologies
Research & Education
Animation
Art
Artificial Intelligence/Machine Learning
Audio
Augmented Reality
Computer Vision
Digital Twins
Education
Ethics and Society
Fabrication
Games
Generative AI
Haptics
Hardware
Image Processing
Modeling
Performance
Pipeline Tools and Work
Real-Time
Scientific Visualization
Simulation
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionLandmine/Garden is an immersive Mixed Reality experience where visitors step into the role of deminers. By collectively searching for concealed landmines, deactivating them, and uncovering the memories and stories buried within the garden, visitors transform a landscape of danger into one of hope, growth, and remembrance.
Art Paper
Arts & Design
Animation
Art
Artificial Intelligence/Machine Learning
Capture/Scanning
Computer Vision
Generative AI
Image Processing
Modeling
Pipeline Tools and Work
Real-Time
Rendering
Full Conference Supporter
Full Conference
Experience
DescriptionCultural heritage sites face threats from climate change, conflict, and degradation. Digital documentation has become a preservation necessity, demonstrated by Notre-Dame's 2019 fire. However, academic photogrammetry research remains disconnected from practical implementation by heritage institutions.
This project demonstrates architectural digitization is achievable using commercial tools. By documenting our methodology we provide professionals with replicable approaches to preserve cultural sites. Computer graphics techniques can directly serve the preservation of cultural heritage.
This project demonstrates architectural digitization is achievable using commercial tools. By documenting our methodology we provide professionals with replicable approaches to preserve cultural sites. Computer graphics techniques can directly serve the preservation of cultural heritage.
Talks
Arts & Design
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Art
Artificial Intelligence/Machine Learning
Audio
Computer Vision
Digital Twins
Education
Fabrication
Games
Generative AI
Geometry
Industry Insight
Modeling
Performance
Pipeline Tools and Work
Real-Time
Rendering
Full Conference Supporter
Full Conference
DescriptionLatentScore turns typed scene descriptions into procedural soundscapes you can steer — running locally on CPU. An open-source Python SDK maps language to structured synth parameters in under a second. See a live audiovisual demo where prompts become sound, palettes, and particles.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Robotics
Simulation
Full Conference Supporter
Full Conference
DescriptionLayerInbetween is an occlusion-aware framework for vector stroke correspondence and automatic inbetweening. It performs automatic layering to guide stroke tracing and correspondence finding for occluded strokes, and to resolve occlusion with layers in the inbetween frames. The vector-based method enables more flexible editing of 2D animation than raster-based video generation.
Computer Animation Festival
Full Conference Supporter
Full Conference
DescriptionA group of ants goes on a field trip into deep underground to explore ancient fossils, but their adventure triggers a collapse that turns them into the next layer of Earth's history.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Generative AI
Geometry
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionWe introduce a fully automated 3D virtual try-on system that fits diverse multilayer garments onto arbitrary humanoid characters. Using SMPL as an intermediate proxy, it handles unrigged, non-manifold assets, supports garment resizing, and produces high-quality draping without manual correspondence, shared rigs, or 2D sewing patterns.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Generative AI
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionA high-quality facial appearance capture method from casual smartphone videos with a powerful delighting prior.
Educator's Forum
Arts & Design
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Animation
Art
Computer Vision
Digital Twins
Education
Games
Modeling
Pipeline Tools and Work
Rendering
Simulation
Spatial Computing
Full Conference Supporter
Full Conference
Experience
DescriptionThis short article introduced our experiences in teaching computer graphics through authentic engineering challenges to student members in an academic makerspace at the University of Hong Kong. It highlights the importance of computer graphics technologies in solving emerging engineering challenges.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionNEO is a neural framework for fast Laplace-Beltrami spectral analysis on 3D point clouds. Rather than computing eigenvectors with expensive iterative solvers, it predicts low-frequency eigenspaces directly from geometry, enabling near-linear scaling, robustness to irregular sampling, and accurate zero-shot transfer to substantially higher-resolution shapes and downstream spectral tasks.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionWe introduce an implicit representation for sewing patterns using distance fields. Combined with latent flow matching, the model learns distributions over panel geometry and stitching, enabling generation, image-based estimation, and optimization. The framework supports flexible topology and improves accuracy over prior methods.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Fabrication
Geometry
Modeling
Full Conference Supporter
Full Conference
DescriptionProblem-aware quad mesh design is challenging due to the choice of neural representation. We propose a two-stage strategy: we first learn sparse singularities, follwed by an analytical step to convert the singularities into cross field and the final quad. Thanks to the hybrid neural-analytical strategy, we can achieve a problem-aware quad style, which is beneficial to downstream animation or simulation.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionWe present a differentiable framework to automatically learn view-dependent 2D kernels in a splatting-based pipeline to improve reconstruction quality and representation efficiency for novel 3D view synthesis.
Emerging Technologies
Arts & Design
Gaming & Interactive
New Technologies
Research & Education
Art
Artificial Intelligence/Machine Learning
Audio
Augmented Reality
Capture/Scanning
Computer Vision
Digital Twins
Display
Dynamics
Education
Fabrication
Games
Generative AI
Graphics Systems Architecture
Haptics
Hardware
Image Processing
Performance
Physical AI
Pipeline Tools and Work
Real-Time
Robotics
Scientific Visualization
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionLentiMorpho is a reconfigurable light field display that dynamically switches between view-dependent and view-independent presentation by varying liquid volume within a glass-bead lens array. By physically enabling or disabling optical functionality, the system allows a single display to alternate between individualized and shared visual experiences without permanent optical trade-offs.
Immersive Pavilion
Arts & Design
Gaming & Interactive
New Technologies
Research & Education
Animation
Art
Artificial Intelligence/Machine Learning
Audio
Augmented Reality
Computer Vision
Digital Twins
Education
Ethics and Society
Fabrication
Games
Generative AI
Haptics
Hardware
Image Processing
Modeling
Performance
Pipeline Tools and Work
Real-Time
Scientific Visualization
Simulation
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionThrough the story of Golnaz, a young Iranian immigrant, LESS THAN 5 GR OF SAFFRON immerses us in the heart of the trauma imposed by forced immigration.
Immersive Pavilion
Arts & Design
Gaming & Interactive
New Technologies
Research & Education
Animation
Art
Artificial Intelligence/Machine Learning
Audio
Augmented Reality
Computer Vision
Digital Twins
Education
Ethics and Society
Fabrication
Games
Generative AI
Haptics
Hardware
Image Processing
Modeling
Performance
Pipeline Tools and Work
Real-Time
Scientific Visualization
Simulation
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionOpVerse demonstrates the complete journey from CT imaging to surgical intervention using XR technology. Attendees experience patient-specific digital twin visualization and hands-on XR-guided surgical planning on a physical phantom, with preliminary clinical results suggesting a ~15-minute reduction in preoperative planning time.
ACM SIGGRAPH 365
Arts & Design
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Animation
Art
Artificial Intelligence/Machine Learning
Audio
Augmented Reality
Capture/Scanning
Computer Vision
Digital Twins
Display
Diversity, Equity and Inclusion
Dynamics
Education
Ethics and Society
Fabrication
Games
Generative AI
Geometry
Graphics Systems Architecture
Haptics
Hardware
Image Processing
Industry Insight
Lighting
Math Foundations and Theory
Modeling
Performance
Physical AI
Pipeline Tools and Work
Real-Time
Rendering
Robotics
Scientific Visualization
Simulation
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionJoin LGBTQIA+ attendees and allies for the Rainbow Meetup, a welcoming space to connect, celebrate identity, and find a queer community at SIGGRAPH. To ensure a safe, respectful environment, the ACM Code of Conduct will be strictly enforced. Harassment or non-consensual behavior, including unauthorized photography, will not be tolerated. Come help build community, find support, and learn about the future of the Rainbow Affinity Group.
Immersive Pavilion
Arts & Design
Gaming & Interactive
New Technologies
Research & Education
Animation
Art
Artificial Intelligence/Machine Learning
Audio
Augmented Reality
Computer Vision
Digital Twins
Education
Ethics and Society
Fabrication
Games
Generative AI
Haptics
Hardware
Image Processing
Modeling
Performance
Pipeline Tools and Work
Real-Time
Scientific Visualization
Simulation
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionYou enter a calm yet unfamiliar world, lying dormant as you arrive. As you explore its surroundings, the world gradually opens up, revealing its light. The more you engage with it, the more you become enveloped in a soothing yet lively awakening, full of surprises.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Fabrication
Generative AI
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionWe present a framework for lifting image-space stylizations into path-space rendering, and demonstrate it on line and tone rendering. By constructing geometric mappings between image and path space, our method preserves line and tone structure under complex light transport, enabling coherent, expressive non-photorealistic rendering within the stylized rendering equation.
Art Paper
Arts & Design
Animation
Art
Artificial Intelligence/Machine Learning
Audio
Capture/Scanning
Computer Vision
Display
Ethics and Society
Generative AI
Hardware
Image Processing
Physical AI
Real-Time
Robotics
Scientific Visualization
Simulation
Spatial Computing
Full Conference Supporter
Full Conference
Experience
DescriptionAs AI becomes an opaque "Black Box," Light Architecture physically "translates" invisible algorithmic structures instead of merely exploiting generative outputs. Synthesizing kinetic light and spatial audio into a Gesamtkunstwerk, this "Experiential AI" offers a sensory alternative to traditional Explainable AI (XAI), providing a critical framework for demystifying the "Hyperobject" of modern computation. Reclaiming agency requires a visceral, embodied encounter with technological systems rather than just intellectual analysis.
Educator's Forum
Arts & Design
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Animation
Art
Augmented Reality
Capture/Scanning
Education
Ethics and Society
Generative AI
Math Foundations and Theory
Modeling
Performance
Pipeline Tools and Work
Real-Time
Rendering
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionThis talk explores Light Field Cinematography through the development of an interdisciplinary Volumetrics Lab supporting faculty-led undergraduate research. Bridging computer science, real-time graphics, cinematography, and embodied performance, it demonstrates how motion-based volumetric capture enables new forms of cinematic, computational, and conceptual problem solving through spatial engagement within real-world educational constraints.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Modeling
Rendering
Simulation
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Technical Paper
Research & Education
Games
Real-Time
Rendering
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionLightOpt is a differentiable optimization framework that automatically reduces the number of lights in real-time game scenes while preserving visual appearance. By optimizing and restructuring real-time light sources, it lowers lighting cost, reduces overlap, and improves cross-platform performance for modern game rendering.
Technical Workshop
Arts & Design
Research & Education
Art
Artificial Intelligence/Machine Learning
Computer Vision
Education
Generative AI
Image Processing
Industry Insight
Rendering
Scientific Visualization
Full Conference Supporter
Full Conference
DescriptionVisual abstraction --- from sketches and diagrams to icons and comics --- enables people to communicate, express, and create by distilling complex visual information into simplified forms that preserve essential meaning. These simplified forms support perception, reasoning, and communication in ways that detailed depiction often cannot. Visual abstraction has been studied across cognitive science, art theory, and computer graphics, offering complementary perspectives on how humans perceive, produce, and interpret such representations.
This workshop brings together these perspectives to examine abstraction as a cognitive mechanism, a communicative tool, and a computational challenge, exploring questions such as: Why and how do people abstract? What makes a simplification effective? How can machines learn to abstract in ways aligned with human perception? And as AI excels at generating realistic images, what role might abstraction play in how humans and AI create together?
This workshop brings together these perspectives to examine abstraction as a cognitive mechanism, a communicative tool, and a computational challenge, exploring questions such as: Why and how do people abstract? What makes a simplification effective? How can machines learn to abstract in ways aligned with human perception? And as AI excels at generating realistic images, what role might abstraction play in how humans and AI create together?
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Generative AI
Geometry
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionWe present an LLM-enhanced scene graph framework for autonomous household rearrangement. By transforming scene graphs into affordance-enhanced graphs that encode scene-grounded object functionality and user preferences, our method detects misplaced objects and plans correct placements. We demonstrate state-of-the-art performance with an end-to-end robotic system in simulated and real-world environments.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionLoBoFit brings robust, high-fidelity garment refitting to challenging avatar changes. It preserves original design features, fine wrinkles, and fit style by optimizing garments in blended local bone coordinates, enabling stable convergence, efficient refinement, and reliable results across large body variations, arbitrary poses, and both single- and multi-layer clothing.
Technical Paper
Research & Education
Geometry
Hardware
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionWe present a GPU system for automatic differentiation for functions defined on triangle meshes that exploits locality and sparsity to deliver efficient derivative computation. From simple user-defined energy terms, the system produces gradients, Jacobians, and sparse Hessians, accelerating simulation and optimization workflows and shifting the computational bottleneck away from differentiation.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Generative AI
Lighting
Full Conference Supporter
Full Conference
DescriptionThis work introduces long-horizon event-based video generation, a challenging task prone to severe temporal inconsistency, and presents LongE2V, a novel approach leveraging video diffusion priors to jointly handle video reconstruction, prediction, and frame interpolation, recovering high-quality videos from sparse event streams.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Augmented Reality
Generative AI
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionLooseRoPE enables prompt-free image editing by letting users crop and paste objects directly into new scenes. It improves diffusion-based harmonization by modulating RoPE with saliency, balancing identity preservation and contextual blending. This yields intuitive, spatially precise edits that maintain object appearance, pose, and placement without requiring text prompts or masks.
Appy Hour
Arts & Design
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Animation
Art
Artificial Intelligence/Machine Learning
Audio
Augmented Reality
Computer Vision
Digital Twins
Education
Ethics and Society
Games
Generative AI
Geometry
Image Processing
Math Foundations and Theory
Modeling
Performance
Physical AI
Pipeline Tools and Work
Real-Time
Robotics
Scientific Visualization
Simulation
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionLost Spirits is an interactive game prototype exploring emotionally resonant local AI characters on edge devices. The player acts as a medium, helping a fractured ghost reconstruct her identity. Built via an AI-augmented workflow, the experience allows players to guide the spirit’s evolution toward final redemption or condemnation.
Computer Animation Festival
Games Summit
Gaming & Interactive
Games
Full Conference Supporter
Full Conference
DescriptionTwo thieves enter the same dungeon searching for a gem worth incredible wealth, but what they find along the way is worth more than all the treasure in the dungeon combined.
Technical Paper
Research & Education
Animation
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionWe present a Koopman operator method for fast deformable simulation using Dynamic Mode Decomposition, enabling efficient long-term prediction without sequential time stepping. Our discretization-agnostic formulation generalizes across shapes and resolutions, making Koopman-based reduced models practical for geometry variation, control, and fast shape optimization tasks.
Technical Paper
Research & Education
Display
Generative AI
Hardware
Image Processing
Full Conference Supporter
Full Conference
DescriptionMaster the night with LUCID: Learning Unified Control for Image Deflaring and Exposure Mastery in Nighttime Photography. Merging flare disentanglement and generative diffusion, our framework conquers entangled degradations. LUCID unlocks continuous control over global illumination and light source scale, enabling the brightening of deep shadows while strictly preventing light pollution.
Technical Paper
Research & Education
Display
Generative AI
Hardware
Image Processing
Full Conference Supporter
Full Conference
DescriptionStage 3 revisions add a DNG-to-sRGB pipeline description and ISP-limitation caveat, promote six supplement figures/tables to the main paper, clarify use of unseen handheld captures for real-world evaluation, added a failure case, limitation and future work in the main paper, and add supplementary motion statistics and a no-motion HDR+ comparison.
Computer Animation Festival
Games Summit
Gaming & Interactive
Games
Full Conference Supporter
Full Conference
DescriptionIn a Solarpunk Universe, Ren and Cora’s peaceful life is shattered by an odd phenomenon spreading through water. They have to go on an adventure to find the source of it and save their ecosystem.
Technical Paper
Research & Education
Animation
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionWe present a multibody simulation framework leveraging affine body dynamics with a co-rotational formulation, enabling pre-factorized system matrices under implicit integration. Mapping primal coordinates onto a dual space via KKT, our method exactly enforces joint constraints, achieving interactive rates for hundreds of thousands of bodies on a single CPU core.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Generative AI
Modeling
Full Conference Supporter
Full Conference
DescriptionMACE-Dance generates dance videos from music using cascaded motion and appearance experts. It first produces expressive 3D dance motion from music, then animates a reference image into a temporally coherent video. The paper also introduces MA-Data, a large-scale dataset and evaluation protocol for this task.
Art Paper
Arts & Design
Animation
Art
Artificial Intelligence/Machine Learning
Audio
Capture/Scanning
Computer Vision
Display
Ethics and Society
Generative AI
Hardware
Image Processing
Physical AI
Real-Time
Robotics
Scientific Visualization
Simulation
Spatial Computing
Full Conference Supporter
Full Conference
Experience
DescriptionMachine Civilization offers SIGGRAPH community a critical lens for examining the gap between explainability and intelligibility in the age of intelligence. This work explored non-human meaning-making and comprehension limits in human-machine interaction. It raises an ethical question: can we engage with machine intelligence as genuine alterity rather than tools? This research proposes that recognizing the alterity of machine intelligence and rethinking de-anthropocentric innovation may be essential for AI's next stage.
Technical Paper
Research & Education
Geometry
Hardware
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionWe present Manifold k-NN, a dynamic programming-based framework for accelerated k-nearest neighbor search on manifold point clouds. By recursively partitioning insertion history via Voronoi successor lists, our method achieves 1–10× speedup over kd-trees in volume-to-surface queries, while supporting zero-overhead prefix queries and dynamic point deletion.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Augmented Reality
Generative AI
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionMAOAM is a unified selection framework that enables precise object- and material- level selection across both text- and click-based interactions. MAOAM leverages a VLM which interprets the user's selection intent, and encodes information to the segmentation head. MAOAM achieves strong performance and generalization across varied objects, materials, and prompting scenarios.
Technical Paper
Research & Education
Geometry
Modeling
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Full Conference Supporter
Full Conference
DescriptionRobust perception benefits from complementary sensors, yet close-range setups remain scarce. We characterize optical and near-field radar imagers via multimodal calibration, collecting data from RGB-D sensors and one mmWave MIMO radar. We evaluate depth sensing across materials, geometries, and distances, release MAROON, and show applications in inverse rendering and reconstruction.
Art Paper
Arts & Design
Art
Computer Vision
Education
Fabrication
Generative AI
Geometry
Hardware
Image Processing
Math Foundations and Theory
Performance
Real-Time
Robotics
Full Conference Supporter
Full Conference
Experience
DescriptionAfter all the presentations, attendees are encouraged to engage in a Q&A session to discuss the topics presented. This will be followed by an Art Papers wrap-up, summarizing key insights, and a preview by the SIGGRAPH 2027 Art Papers chair, offering a glimpse into the upcoming 2027 Art Papers program.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Modeling
Real-Time
Rendering
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Full Conference Supporter
Full Conference
DescriptionThis paper tackles the task of learning to generate signals over triangle meshes in a triangulation-agnostic manner, meaning the trained model can be applied to different mesh triangulations effectively. It proposes a specific noise distribution which is triangulation agnostic on meshes, to be used inside the Flow Matching's denoising process.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Games
Generative AI
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionMAViD is a multimodal framework for interactive audio-visual dialogue. Its Conductor-Creator architecture integrates multimodal understanding and reasoning into response generation. By combining AR and diffusion models, MAViD enables synchronized, long-duration audio-video generation with consistent identity and timbre, delivering vivid and contextually coherent digital human interactions.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Fabrication
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionWe explore a new direction in mechanical cloaking: halftoning an image using a porous structure that behaves like a linear, isotropic material and visually matches an image. For an external observer, this creates the surprising effect where the object appears mechanically homogeneous while its porous structure resembles a target image.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Generative AI
Geometry
Modeling
Rendering
Simulation
Full Conference Supporter
Full Conference
DescriptionMegaNorm is a fast, accurate, and robust method for orienting normals in non-watertight, scene-scale point clouds, scaling to ~50M points in under 10 minutes. It uses two feed-forward networks: one orients local patches in parallel, while a lightweight MLP evaluates pairwise consistency. Finally, a 0–1 integer program enforces global consistency.
Art Paper
Arts & Design
Animation
Art
Artificial Intelligence/Machine Learning
Capture/Scanning
Computer Vision
Generative AI
Image Processing
Modeling
Pipeline Tools and Work
Real-Time
Rendering
Full Conference Supporter
Full Conference
Experience
DescriptionAfter all the presentations, attendees are invited to participate in a Q&A.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Fabrication
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionWe present a high-performance Newton-type solver designed for the nonlinear finite element problems common in simulation and geometry processing. By leveraging the inherent block structure of spatially discretized systems across all solver phases, we achieve significant speedups over existing direct-solver-based frameworks without compromising accuracy, flexibility, or ease of use.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Full Conference Supporter
Full Conference
DescriptionMeshFlow is a powerful diffusion-based framework utilizing equivariant modeling to transform random triangle soups into high-quality meshes in less than one second.
Computer Animation Festival
DescriptionMetachess is a generative system that translates chess games into spatial structures. Using PGN chess notation as input, parses each move and transforms strategic decisions into a growing architectural form. As the game unfolds, moves become trajectories, accumulating into a three-dimensional structure.
The second work visualizes the historic 1997 rematch between world chess champion Garry Kasparov and IBM’s Deep Blue, a milestone in the history of artificial intelligence. Through the Metachess system, the game is no longer represented as notation or replay, but as a spatial trace of strategy emerging in time.
Each game produces a unique structure that can also be fabricated through metal additive manufacturing. The system can operate in real time, translating live matches into evolving forms. Metachess proposes a new way of reading games as architectural patterns of decision-making, where players shift from opponents to co-creators of a shared spatial artifact.
The second work visualizes the historic 1997 rematch between world chess champion Garry Kasparov and IBM’s Deep Blue, a milestone in the history of artificial intelligence. Through the Metachess system, the game is no longer represented as notation or replay, but as a spatial trace of strategy emerging in time.
Each game produces a unique structure that can also be fabricated through metal additive manufacturing. The system can operate in real time, translating live matches into evolving forms. Metachess proposes a new way of reading games as architectural patterns of decision-making, where players shift from opponents to co-creators of a shared spatial artifact.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionWe present Mix3R, a method that mixes feed-forward reconstruction and generative 3D priors using a Mixture-of-Transformer architecture for multi-view aligned generative 3D reconstruction from sparse views.
Technical Paper
Research & Education
Animation
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionOur variant of the Material Point Method discretizes both the velocity and stress fields over mixed finite elements, avoiding the need for repeated costly particle-to-grid transfers within the implicit elastoviscoplastic solve. Our method supports materials ranging from sand and snow to elastic solids, with two-way coupling to rigid-body solvers.
Educator's Forum
Gaming & Interactive
New Technologies
Research & Education
Augmented Reality
Digital Twins
Education
Games
Image Processing
Real-Time
Scientific Visualization
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionThis Bayer Filter assignment introduces digital image formation concepts in undergraduate Computer Graphics courses while playing with different components of light and color. Students get visual feedback as they take a deep, step-by-step dive into working with digital image data structures combining flashlights, worksheets, code, and photographs.
Technical Paper
Research & Education
Animation
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionMixwell introduces sharp 2D fluid brushes and GPU-accelerated analytical methods for progressive, resolution-independent physics-based mixing. Derived from potential flow around cylindrical tines, Mixwell evaluates drift per sample without grids or intermediate resampling, enabling real-time, arbitrary-resolution fluid mixing and rendering with negligible numerical dissipation.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Real-Time
Rendering
Full Conference Supporter
Full Conference
DescriptionWe propose Mobile3DGS³, the first mobile 3DGS acceleration paradigm solving high-res real-time rendering pain points via gradient-assisted post-rendering. It features TSGC, GASS, GAFI for alternating super-sampling/interpolation, delivering high-fidelity efficiency.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionOur method enhances noisy multi-human object interaction (MHOI) data by automatically generating plausible hand and finger motions and refining noisy body movements, all while preserving the original interaction semantics.
Talks
Arts & Design
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Animation
Artificial Intelligence/Machine Learning
Augmented Reality
Capture/Scanning
Computer Vision
Digital Twins
Games
Geometry
Hardware
Image Processing
Performance
Pipeline Tools and Work
Rendering
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionThe talk covers transitioning to a render dependency graph implementation with automatic memory management, GPU synchronization, and async compute scheduling, along with a rich suite of debug tools — including a real-time frame analysis tool. Case studies include raytracing acceleration structures and terrain virtual texturing.
Computer Animation Festival
Games Summit
Gaming & Interactive
Games
Full Conference Supporter
Full Conference
DescriptionA story of early parenthood that preys on the fears of new parents. Someone is always watching…
Monitor is an independent short film from Michael Berardini, a Director of Scrat Tales and Batwheels. The short is done in a unique style blending 2D characters in a 3D environment to give the entire film an illustrative quality. Batu Sener (Scrat Tales, Better Man) delivers a beautiful but haunting score while Jérémy Ben Ammar provides a quietly creepy soundscape.
Monitor is an independent short film from Michael Berardini, a Director of Scrat Tales and Batwheels. The short is done in a unique style blending 2D characters in a 3D environment to give the entire film an illustrative quality. Batu Sener (Scrat Tales, Better Man) delivers a beautiful but haunting score while Jérémy Ben Ammar provides a quietly creepy soundscape.
Technical Paper
Research & Education
Geometry
Modeling
Rendering
Simulation
Full Conference Supporter
Full Conference
DescriptionThis paper extends Monte Carlo PDE solvers to handle nonlinear radiative boundary conditions using a Picard-style fixed-point iteration framework. It introduces a heteroscedastic regression-based denoising method for boundary estimates and demonstrates the approach on heat radiation simulations with complex geometries, achieving higher accuracy than standard linearization strategies.
Technical Paper
Research & Education
Geometry
Hardware
Modeling
Real-Time
Rendering
Full Conference Supporter
Full Conference
DescriptionBiharmonic diffusion curves are lightweight vector graphics primitives that offer artistic control over color gradients. We present their first Monte Carlo renderer, which builds upon recent advances in Monte Carlo geometry processing. We phrase the biharmonic problem as two coupled second-order problems and recover boundary data using variational energy minimization.
Technical Paper
Research & Education
Generative AI
Geometry
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionMoonSplat is a robust and efficient online voxelized 3DGS reconstruction framework integrated with global Sim(3) optimization, which enables reliable camera tracking and efficient global loop closure for both camera poses and voxelized 3DGS. MoonSplat is equipped with a color residual learning strategy for real-time online optimization of voxelized 3DGS.
Technical Paper
Research & Education
Augmented Reality
Display
Fabrication
Hardware
Real-Time
Rendering
Simulation
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionShape-changing displays typically lose pixel density as surface area expands, limiting usability. We introduce MorphSkein, a shape-changing afterimage display that preserves a minimum density (1.44 px/cm in our prototype) across naturally occurring axisymmetric shapes of varying surface area generated by spinning cables (troposkeins).
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Robotics
Simulation
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Robotics
Simulation
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Robotics
Simulation
Full Conference Supporter
Full Conference
DescriptionThis work proposes a novel framework, Motion4Motion, for transferring a motion from one subject to another in a video. Without a skeleton or neural network training, Motion4Motion runs efficiently at inference.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Robotics
Simulation
Full Conference Supporter
Full Conference
DescriptionMotionBricks is a real-time generative framework that transforms interactive motion control for animation and robotics. By combining a large-scale latent backbone with intuitive "smart primitives," it delivers high-quality, zero-shot motion synthesis at 15,000 FPS, allowing users to effortlessly build complex and fine-grained animations and robotic applications.
Emerging Technologies
Arts & Design
Gaming & Interactive
New Technologies
Research & Education
Art
Artificial Intelligence/Machine Learning
Audio
Augmented Reality
Capture/Scanning
Computer Vision
Digital Twins
Display
Dynamics
Education
Fabrication
Games
Generative AI
Graphics Systems Architecture
Haptics
Hardware
Image Processing
Performance
Physical AI
Pipeline Tools and Work
Real-Time
Robotics
Scientific Visualization
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionWe present a non-wearable VR system that forms stereoscopic aerial images in front of the user. By eliminating head-mounted devices, our mountless aerial HMD enables quick, comfortable transitions between physical and virtual spaces, offering a new design space for immersive VR experiences.
Immersive Pavilion
MoveTogether: Exploring Physical Co-op Gameplay in Co-located Mixed Reality via Modular Shared Props
10:30am - 5:00pm PDT Monday, 20 July 2026 Concourse HallArts & Design
Gaming & Interactive
New Technologies
Research & Education
Animation
Art
Artificial Intelligence/Machine Learning
Audio
Augmented Reality
Computer Vision
Digital Twins
Education
Ethics and Society
Fabrication
Games
Generative AI
Haptics
Hardware
Image Processing
Modeling
Performance
Pipeline Tools and Work
Real-Time
Scientific Visualization
Simulation
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionMoveTogether, a novel physical co-op gameplay in which two players manipulate a single tracked prop, creating a shared physical communication channel that complements visual and audio cues. We showcase the modular prop kit MingleKit through the game TimberLand, a sequence of short cooperative challenges that encourages players to explore coordination.
Course
Gaming & Interactive
Artificial Intelligence/Machine Learning
Games
Hardware
Lighting
Performance
Real-Time
Rendering
Full Conference Supporter
Full Conference
DescriptionA technical introduction to mobile graphics spanning the spectrum of hardware, software and developer experience, exploring the state of the art. We look at how neural graphics, GPU-driven rendering, global illumination and ray tracing techniques are being brought to mobile.
Technical Paper
Research & Education
Fabrication
Geometry
Modeling
Robotics
Simulation
Full Conference Supporter
Full Conference
DescriptionThis work presents a new class of gear mechanisms, called multi-point conjugation gear mechanisms, as well as computational techniques to modeling these gear mechanisms for exactly generating user-specified 3D motions under external loads.
Technical Paper
Research & Education
Animation
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionMPM Lite is a hybrid Lagrangian/Eulerian method that removes particle-based quadrature at solve time. It achieves 15.9 times speedup over implicit MPM, and 1.88 times speedup over explicit MPM.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Rendering
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionMTPano is a robust multi-task foundation model for panoramic scene understanding. Trained via a label-free pipeline using perspective dense priors, it adopts a dual-stream architecture that resolves feature conflicts between depth, segmentation, and surface normal estimations, achieving state-of-the-art performance on spherical images.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Lighting
Modeling
Real-Time
Rendering
Full Conference Supporter
Full Conference
DescriptionWe propose MRBNN that delivers real-time photorealistic volume rendering. We leverage reformulation of in-scattering integral and design an efficient neural representation and sampling strategy. Our method achieves rendering in a few milliseconds with the ability to handle complex cases such as spatially varying albedo and heterogeneous phase.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Real-Time
Rendering
Full Conference Supporter
Full Conference
DescriptionThis work addresses disocclusion noise in ReSTIR by extending the framework to multiple screen-space layers. This enables occluded samples in previous frames to contribute to the final image, improving temporal reuse in disoccluded regions. Depth ranges and active domains are introduced to reduce computational overhead while maintaining quality.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Robotics
Simulation
Full Conference Supporter
Full Conference
DescriptionText-to-motion models struggle with composite prompts, often collapsing to a single action. We present MultiAct, an inference-time, unpaired framework that enhances pretrained generators by amplifying cross-attention for underrepresented tokens. A lightweight decision scheme selects optimal parameters per prompt, improving semantic coverage while preserving motion realism without retraining or architectural changes.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Games
Generative AI
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionWe present a data-driven framework for physics-based, muscle-driven dexterous control that enables musculoskeletal hands to perform piano playing for novel musical scores. Our framework adopts a hierarchical architecture, combining high-frequency muscle control with low-frequency coordination. The method achieves coordinated bimanual motions and produces human-like activation patterns, generalizing beyond reference data.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Generative AI
Rendering
Full Conference Supporter
Full Conference
DescriptionMV-S2V is the first work to explore video generation with multi-view subject consistency. It presents a simple yet effective framework, including a tailored data curation pipeline and a Temporally-Shifted RoPE (TS-RoPE) specially designed for multi-view reference conditioning, achieving superior 3D subject consistency w.r.t. multi-view reference images and high-quality visual outputs.
Technical Paper
Research & Education
Fabrication
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionWe introduce NeuBase, a neural parametric surface representation that both accurately fits target surfaces with fine geometric detail and supports intuitive real time surface deformation.
Technical Paper
Research & Education
Fabrication
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionWe present Neural Piecewise Parametric Surfaces (NeuPPS) that models complex surface geometries with a coarse patch layout containing arbitrary n-sided surface patches, offering enhanced flexibility, high fitting precision, and ensured continuity between adjacent patches. Applications to surface fitting and shape space learning tasks demonstrate its advantages over existing methods.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Rendering
Robotics
Full Conference Supporter
Full Conference
DescriptionWe pair Neural Cellular Automata with a lightweight Neural Field that decodes coarse cell states and local coordinates into high-resolution appearance, decoupling lattice size from output resolution. Trained end-to-end with task-specific losses, our model synthesizes textures and grows morphologies at arbitrary resolution across 2D/3D/mesh in real-time while preserving self-organizing behavior.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Rendering
Robotics
Full Conference Supporter
Full Conference
DescriptionNeural Particle Automata extend Neural Cellular Automata from fixed grids to dynamic particles. Each particle carries a continuous position and internal state, updated by a shared neural rule with SPH-based local perception. Backed by optimized CUDA kernels, NPA learn scalable self-organizing dynamics for morphogenesis, particle texture synthesis, and point-cloud classification.
Technical Paper
Research & Education
Geometry
Modeling
Rendering
Simulation
Full Conference Supporter
Full Conference
DescriptionLearnable adaptive sampler and integrator using Transformer.
Technical Paper
Research & Education
Geometry
Hardware
Modeling
Real-Time
Rendering
Full Conference Supporter
Full Conference
DescriptionWe introduce NeuralSketch2Surf, the first fast and robust neural surfacing solution, capable of processing arbitrary unoriented sketches at interactive rates.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Full Conference Supporter
Full Conference
DescriptionNexus is a revolutionary two-stage diffusion framework for generating high-fidelity 3D polygon meshes. By replacing slow sequential modeling with geometry structure and topology diffusion, Nexus effortlessly handles complex, non-manifold geometries and achieves more controllable and faster mesh generation. It delivers significantly superior artistic meshes compared to current state-of-the-art methods.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionWe introduce a novel modular modular meta-sampler architecture that bridges local subspace and global sampling, allowing integrator designers to employ specialized low-dimensional samplers while still achieving high-dimensional uniformity.
Educator's Forum
Gaming & Interactive
Research & Education
Education
Games
Image Processing
Modeling
Real-Time
Full Conference Supporter
Full Conference
Experience
DescriptionThis talk discusses the course philosophy, design, and implementation details of the undergraduate "Video Game Programming" course at the University of Southern California, which has students program both 2D and 3D games largely from scratch.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Full Conference Supporter
Full Conference
DescriptionWe present a LOS-guided NLOS imaging pipeline for arbitrary relay surface geometries, representing hidden scenes with 3D Gaussian primitives and developing an efficient differentiable transient renderer. Real experiments on complex relay surfaces captured with our custom system show improved robustness and broaden the practical applicability of NLOS imaging.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Fabrication
Generative AI
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Fabrication
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Technical Paper
Object-Space Analysis of Local Contrast Sensitivity for Hierarchical Representations of 3D Gaussians
2:10pm - 2:20pm PDT Thursday, 23 July 2026 Room 408 AResearch & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionWe present an object-space contrast sensitivity analysis for 3D Gaussian representations that estimates spatial frequency responses without explicit rasterization using analytical Fourier transforms and the projection-slice theorem. This enables contrast-sensitivity-based perceptual evaluation and efficient foveated rendering via hierarchical Gaussian level selection, preserving visual quality while reducing rendering cost.
Computer Animation Festival
Games Summit
Gaming & Interactive
Games
Full Conference Supporter
Full Conference
DescriptionOcean Futurisms is a collaborative project between TBA21–Academy, Radical Ocean Futures and Inferstudio. Through three original animated short films, the project invites audiences to step into alternative, science-based futures for the ocean. Blending scientific forecasts with immersive storytelling, each of the films rehearses questions for the future of ocean policy/governance through speculative scenario making and worldbuilding. These films act as simulated visions of worlds that might yet be…for better, or for worse.
In Havsrå, a direct action activist organisation mobilises a global audience by live-streaming their operations as they attempt to disrupt illegal deep sea mining.
Designed for use in workshops and educational screenings with ocean scientists and policymakers, the Ocean Futurisms films are an ambitious experiment in the power of narrative to inspire real-world conversations about the radical and urgent changes required to safeguard Earth's oceans for the future.
In Havsrå, a direct action activist organisation mobilises a global audience by live-streaming their operations as they attempt to disrupt illegal deep sea mining.
Designed for use in workshops and educational screenings with ocean scientists and policymakers, the Ocean Futurisms films are an ambitious experiment in the power of narrative to inspire real-world conversations about the radical and urgent changes required to safeguard Earth's oceans for the future.
Computer Animation Festival
DescriptionOcean Futurisms is a collaborative project between TBA21–Academy, Radical Ocean Futures and Inferstudio. Through three original animated short films, the project invites audiences to step into alternative, science-based futures for the ocean. Blending scientific forecasts with immersive storytelling, each of the films rehearses questions for the future of ocean policy/governance through speculative scenario making and worldbuilding. These films act as simulated visions of worlds that might yet be…for better, or for worse.
In this first film of the project, M.A.N.A, we follow a future ocean noise pollution data analyst - working within an alliance of island states redefining ocean governance amidst the realities of climate crisis.
Designed for use in workshops and educational screenings with ocean scientists and policymakers, the Ocean Futurisms films are an ambitious experiment in the power of narrative to inspire real-world conversations about the radical and urgent changes required to safeguard Earth's oceans for the future.
In this first film of the project, M.A.N.A, we follow a future ocean noise pollution data analyst - working within an alliance of island states redefining ocean governance amidst the realities of climate crisis.
Designed for use in workshops and educational screenings with ocean scientists and policymakers, the Ocean Futurisms films are an ambitious experiment in the power of narrative to inspire real-world conversations about the radical and urgent changes required to safeguard Earth's oceans for the future.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Generative AI
Geometry
Modeling
Simulation
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionOmniHands is a universal architecture for hand reconstruction across diverse scenarios, supporting single-hand reconstruction, two-hand interaction reconstruction, temporal inputs, and multi-view inputs. It achieves this within a unified framework by first tokenizing the features of each hand and then fusing the tokens in a task-dependent manner
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Generative AI
Rendering
Full Conference Supporter
Full Conference
DescriptionThis work introduces a controllable panoramic video generation framework designed for long-horizon scene wandering. It utilizes a global-to-local preview-and-refine pipeline to synthesize high-fidelity, spatially and temporally coherent panoramic videos from a single input panoramic image or video.
Talks
Arts & Design
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Animation
Artificial Intelligence/Machine Learning
Augmented Reality
Capture/Scanning
Computer Vision
Digital Twins
Games
Geometry
Hardware
Image Processing
Performance
Pipeline Tools and Work
Rendering
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionLAIKA’s VFX look development artists create both digital doubles and CG assets that integrate into the artisan-made world of stop motion film. Our tools for on-set visual data acquisition drive look development, validate the congruity of the asset, and produce shot ready assets that fit LAIKA’s bespoke, handmade vision.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Generative AI
Modeling
Full Conference Supporter
Full Conference
DescriptionModern T2I DiTs are powerful yet often lack diversity. We identify a novel intervention point within DiTs: the Contextual Space at the transformer block level, where textual tokens are enriched by image content. We propose a method that repels a batch of samples in this space to produce diverse generations.
Computer Animation Festival
Full Conference Supporter
Full Conference
DescriptionOniko is a dark, stylised animated short set in Edo-period Japan, where a princess is forced to flee her snowbound castle with her newborn as samurai hunt her through the night. Driven into a storm-lashed forest and toward a cliff above the ocean, she must rely on determination, sacrifice, and a terrifying secret bound to her child to survive. Blending historical drama with supernatural myth, Oniko is a tense, atmospheric tale about motherhood, defiance, and the cost of choosing one’s own fate.
Computer Animation Festival
Full Conference Supporter
Full Conference
DescriptionThree neighbors—Eunwoo, a new resident; Bumsaeng, living a quiet senior life; and Jisu, worn out by her routine—seemed to have no reason to connect. Yet, they each open a "door," slowly weaving themselves into one another's worlds.
Course
Research & Education
Geometry
Math Foundations and Theory
Simulation
Full Conference Supporter
Full Conference
DescriptionThis course on Fluid Simulation focuses on Optimal Transport, a mathematical toolbox for creating a 3D mesh representation that follows the movement of the fluid while accurately ensuring volume conservation, fluid-boundary interactions and free-surface evolution. For a hands-on experience, it includes many bring-back-home programming examples, from simple to elaborate.
Course
New Technologies
Animation
Art
Artificial Intelligence/Machine Learning
Audio
Computer Vision
Digital Twins
Generative AI
Rendering
Full Conference Supporter
Full Conference
DescriptionThis course introduces a unified paradigm for multimodal control of joint audio-visual synthesis using the LTX-2 foundation model. Beyond text prompts, we demonstrate precise control via audio, video, imagery, and spatial data. Attendees will learn dual-stream transformer architectural innovations and practical techniques for orchestrating high-fidelity, aligned multimodal generation
Computer Animation Festival
Full Conference Supporter
Full Conference
DescriptionDuring the Hungry Ghost Festival in Hong Kong, 18 year old Jane is on a night out with her friends. When she spots her six year old sister in the busy crowd, she starts a frantic search through the party district that will bring her to face the ghosts of her repressed past.
Spatial Storytelling
Gaming & Interactive
New Technologies
Capture/Scanning
Digital Twins
Generative AI
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionOUT OF THE ASHES co-directors Nonny de la Peña and Rory Mitchell detail the process of creating the work-in-progress immersive documentary beginning with 3D scanning the devastation in the aftermath of the Los Angeles Fires of 2025 through the conception and execution of 3DGS & 4DGS pipelines for Unity.
Technical Paper
Research & Education
Augmented Reality
Display
Fabrication
Hardware
Real-Time
Rendering
Simulation
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionWe present PAColorHolo, a perceptually-aware color management framework that enables accurate color reproduction in holographic displays. Our approach jointly addresses system-level color distortions through color space transformation, adaptive illumination control, and a color-restoration neural network. Simulations, optical experiments, and a user study demonstrate improved perceptual color fidelity.
Technical Paper
Research & Education
Fabrication
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Talks
Production & Animation
Animation
Artificial Intelligence/Machine Learning
Computer Vision
Graphics Systems Architecture
Image Processing
Lighting
Performance
Pipeline Tools and Work
Rendering
Full Conference Supporter
Full Conference
DescriptionNetflix Animation Studios presents Pashboard, a cloud-native evolution of the Rez package manager. By integrating with Netflix’s streaming infrastructure and zero-trust security, Pashboard solves the latency and access hurdles of traditional on-premise pipelines. This novel system enables a high-performance, globally distributed software workflow for artists and developers across six continents.
Talks
Arts & Design
Production & Animation
Research & Education
Animation
Augmented Reality
Dynamics
Games
Geometry
Lighting
Math Foundations and Theory
Modeling
Pipeline Tools and Work
Real-Time
Rendering
Simulation
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionWalt Disney Animation Studios' “Zootopia 2” was the first film to adopt Disney's Hyperion Renderer's second-generation path guiding system. Building on lessons we previously shared in “Path Guiding in Production and Recent Advancements”, this talk presents new advancements of our system that enabled large-scale adoption, improving workflows and rendering efficiency.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionWe present PatternGSL, a structured and template-free representation for simulation-ready 3D garments. Given a single garment image, our VLM-based method predicts editable sewing patterns with explicit panel, edge, and stitch topology. Trained on the large-scale PatternGSLData dataset, it enables accurate reconstruction, cloth simulation, and downstream garment design workflows.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Full Conference Supporter
Full Conference
DescriptionPEGAsus enables personalized 3D shape generation by learning reusable geometry and appearance concepts from a single reference shape, supporting both global and region-wise personalization. The learned concepts can be combined with text to synthesize novel personalized 3D assets outperforming existing state-of-the-art solutions.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Augmented Reality
Rendering
Simulation
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Art Paper
Arts & Design
Research & Education
Animation
Art
Augmented Reality
Dynamics
Hardware
Lighting
Performance
Real-Time
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionAfter all the presentations, attendees are invited to participate in a Q&A.
Games Summit
Gaming & Interactive
Production & Animation
Games
Full Conference Supporter
Full Conference
Experience
DescriptionFilm-grade virtual production workflows are increasingly finding their way onto game MoCap stages — but adapting them to the pace, scale, and constraints of AAA production is its own challenge. This talk details how EA Create Capture met that challenge for Battlefield 6: from 1-to-1 physical set construction and moving vehicle builds to real-time visualization that enabled key creative decisions on set and same-day editorial delivery.
Technical Workshop
Gaming & Interactive
Research & Education
Artificial Intelligence/Machine Learning
Display
Games
Graphics Systems Architecture
Hardware
Performance
Real-Time
Rendering
Full Conference Supporter
Full Conference
DescriptionThe Performance, Rendering, and Interaction for Competitive Esports (PRICE) workshop brings together cutting-edge research on topics essential to esports, including graphics and rendering, systems and performance, human factors and HCI, computer supported cooperative work (CSCW), data, and artificial intelligence.
Esports is no longer a niche hobby; it is a major cultural and economic force, with a global audience in the hundreds of millions and a market size exceeding $1 billion. The performance and experience of both players and spectators are critically dependent on the underlying technology. Success in competitive games hinges on a complex interplay between hardware capabilities, software optimizations, and the limits of human perception, performance, and experience.
The program includes a keynote, sketch presentations, and interactive panel discussions among sketch authors where the audience will be able to participate in the discussion. Additionally, a few short breaks in the agenda will allow for small group and one-on-one interactions.
Technical Workshop Sketches:
AI Level of Detail: Distance-Aware ML Model Precision Selection for Real-Time Human Motion Prediction in Games
Mathew Varghese, University of Washington
Towards Field Experiments in Esports Competition
Arjun Madhusudan, North Carolina State University
Josef Spjut, Ben Boudaoud, Joohwan Kim, NVIDIA
Benjamin Watson, North Carolina State University
Understanding Emergent Non-Verbal Communication in the Delta Force Competitive Video Game through Multimodal AI Analysis
Jinyuan Guo, Duke Kunshan University
Josef Spjut, NVIDIA
Improving Interaction by Transforming the Render-Display Pipeline
Aaron Fulmer, Francis Marion University; North Carolina State University
Evan Jonson, North Carolina State University
Hee-Jin Choi, Sejong University
Hyoungsik Nam, Kyung Hee University
Benjamin Watson, North Carolina State University
What Do Competitive Players Do With Their Settings? A Multi-Game Esports Configuration Survey
Ashish Rajpurohit, Arjun Madhusudan, Benjamin Watson, North Carolina State University
Esports is no longer a niche hobby; it is a major cultural and economic force, with a global audience in the hundreds of millions and a market size exceeding $1 billion. The performance and experience of both players and spectators are critically dependent on the underlying technology. Success in competitive games hinges on a complex interplay between hardware capabilities, software optimizations, and the limits of human perception, performance, and experience.
The program includes a keynote, sketch presentations, and interactive panel discussions among sketch authors where the audience will be able to participate in the discussion. Additionally, a few short breaks in the agenda will allow for small group and one-on-one interactions.
Technical Workshop Sketches:
AI Level of Detail: Distance-Aware ML Model Precision Selection for Real-Time Human Motion Prediction in Games
Mathew Varghese, University of Washington
Towards Field Experiments in Esports Competition
Arjun Madhusudan, North Carolina State University
Josef Spjut, Ben Boudaoud, Joohwan Kim, NVIDIA
Benjamin Watson, North Carolina State University
Understanding Emergent Non-Verbal Communication in the Delta Force Competitive Video Game through Multimodal AI Analysis
Jinyuan Guo, Duke Kunshan University
Josef Spjut, NVIDIA
Improving Interaction by Transforming the Render-Display Pipeline
Aaron Fulmer, Francis Marion University; North Carolina State University
Evan Jonson, North Carolina State University
Hee-Jin Choi, Sejong University
Hyoungsik Nam, Kyung Hee University
Benjamin Watson, North Carolina State University
What Do Competitive Players Do With Their Settings? A Multi-Game Esports Configuration Survey
Ashish Rajpurohit, Arjun Madhusudan, Benjamin Watson, North Carolina State University
Technical Paper
Research & Education
Geometry
Modeling
Full Conference Supporter
Full Conference
DescriptionWe developed a new algorithm to simplify the topology of a 3D shape to user-prescribed numbers of connected components, handles, and voids, while minimizing geometric changes.
Technical Paper
Research & Education
Animation
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Modeling
Rendering
Simulation
Full Conference Supporter
Full Conference
DescriptionWe present a system for optimizing spacecraft designs under radiation pressure, the force that photons exert on objects. It combines a fast Monte Carlo simulation, a neural proxy of it, and gradient-based optimization to find spacecraft geometry, or operational parameters that minimize travel time or maximize proximity to a target.
Course
Arts & Design
Gaming & Interactive
Research & Education
Animation
Artificial Intelligence/Machine Learning
Dynamics
Full Conference Supporter
Full Conference
DescriptionThrough theory and reference code, attendees will learn the simple foundations of elastic simulation. Once the basics are established, the course further explores through the current research landscape of neural physics simulation, showing how newcomers can integrate modern physics simulation into their projects.
Technical Paper
Research & Education
Animation
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionApplying texture maps to dynamic deformable surfaces presents a significant challenge, due to ever-shifting differences in length scales involved.
We present two novel wave-based procedural texturing algorithms which reproduce common physical properties like advection and self-similarity, enabling physics-guided animation of deforming objects with extreme texture map distortions.
We present two novel wave-based procedural texturing algorithms which reproduce common physical properties like advection and self-similarity, enabling physics-guided animation of deforming objects with extreme texture map distortions.
Emerging Technologies
Arts & Design
Gaming & Interactive
New Technologies
Research & Education
Art
Artificial Intelligence/Machine Learning
Audio
Augmented Reality
Capture/Scanning
Computer Vision
Digital Twins
Display
Dynamics
Education
Fabrication
Games
Generative AI
Graphics Systems Architecture
Haptics
Hardware
Image Processing
Performance
Physical AI
Pipeline Tools and Work
Real-Time
Robotics
Scientific Visualization
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionWe present a compact pinlight AR display that achieves a wide field of view and expanded eyebox via dynamic eye tracking. A camera rig captures multi-view pupil images to estimate 3D pupil position using a neural network, enabling more precise 3D eye tracking. A prototype validates the proposed approach.
ACM SIGGRAPH 365
Arts & Design
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Animation
Artificial Intelligence/Machine Learning
Augmented Reality
Capture/Scanning
Computer Vision
Digital Twins
Education
Ethics and Society
Games
Generative AI
Graphics Systems Architecture
Industry Insight
Real-Time
Rendering
Robotics
Scientific Visualization
Simulation
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionComputer Graphics legend Don Greenberg is the ACM/SIGGRAPH Pioneers Featured Speaker for 2026! Greenberg's astonishingly multi-disciplinary career started in architecture, and has spanned the computer graphics revolution from its genesis in the 1960s. He has taught in the fields of architecture, engineering and computer science, while also being an elite athlete and life-long advisor to the tennis and soccer teams at Cornell. Greenberg’s teaching has covered computer-aided design in architecture as well as computer animation in the Department of Art, and he has simultaneously taught courses in computer science and technology strategy. Greenberg received the 1987 ACM SIGGRAPH Steven A. Coons Award for Outstanding Creative Contributions to Computer Graphics, its highest honor, and the 1989 NCGA Academic Award, the highest educational award given by the National Computer Graphics Association. He is a member of the National Academy of Engineering; Founding Fellow, American Institute of Medical and Biological Engineering; Fellow, Association for Computer Machinery; Fellow, American Association for the Advancement of Science; and the founding director of the 11-year National Science and Technology Center for Computer Graphics and Scientific Visualization. His talk will start with a career retrospective, then move to his current work and the issues he feels are most important to be discussed at SIGGRAPH right now in 2026!
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Full Conference Supporter
Full Conference
DescriptionPixal3D is a pixel-aligned paradigm that solves the fidelity bottleneck in image-to-3D synthesis. By utilizing a back-projection conditioning scheme to explicitly lift 2D features into view-aligned 3D volumes, it eliminates ambiguity, enabling scalable generative reconstruction of highly accurate 3D assets and scenes from single or multi-view images.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Generative AI
Lighting
Full Conference Supporter
Full Conference
DescriptionPixel Cube is a novel lighting system for realistic lighting emulation and high-speed video relighting data acquisition. With a high-quality lighting-aware dataset, we train a diffusion model for dynamic portrait relighting. Our model produces photorealistic, temporally consistent portrait videos under new lighting conditions that faithfully preserve the subject's facial features.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Generative AI
Geometry
Modeling
Rendering
Simulation
Full Conference Supporter
Full Conference
DescriptionPixels2Peaks automates the creation of 3D bare-earth terrain heightmaps from a single 2D photograph. Our method first estimates camera parameters and visible geometry; then uses a guided diffusion model to plausibly synthesize occluded regions. The result maintains geomorphological, hydrological, and view consistencies.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Rendering
Robotics
Full Conference Supporter
Full Conference
DescriptionWe propose PixTex, the first pixel-space multi-view diffusion framework for texture generation. By bypassing latent compression and utilizing a coarse-to-fine strategy, it ensures lossless geometric guidance and superior pixel-level multi-view consistency, achieving state-of-the-art, high-fidelity, and seamless results.
Talks
Arts & Design
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Art
Artificial Intelligence/Machine Learning
Audio
Computer Vision
Digital Twins
Education
Fabrication
Games
Generative AI
Geometry
Industry Insight
Modeling
Performance
Pipeline Tools and Work
Real-Time
Rendering
Full Conference Supporter
Full Conference
DescriptionAstral Planes replaced loop-based playback with a Unity-native runtime music system where gameplay reshapes composition directly. This production case study explains tracks, bins, notes, shared timing, persistent loops, and how audio architecture changes when composition becomes part of interaction.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Games
Generative AI
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionPointLLM-R introduces chain-of-thought reasoning for 3D point cloud understanding via a data-centric pipeline. It builds a large reasoning dataset (PoCoTI) and trains a multimodal model that achieves state-of-the-art performance, improving accuracy, interpretability, and robustness across synthetic and real-world 3D tasks.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Full Conference Supporter
Full Conference
DescriptionSigned distance computation has been either fast or robust, but not both. We explain why, using a theory that unifies signed distance with reconstruction. Our method directly estimates signed distance from point clouds, providing fast pointwise evaluation at arbitrary spatial resolution, without requiring discretization, global optimization, or explicit reconstruction.
Course
Research & Education
Animation
Dynamics
Geometry
Hardware
Image Processing
Math Foundations and Theory
Modeling
Rendering
Simulation
Full Conference Supporter
Full Conference
DescriptionLike a semester long graduate seminar on Poisson Problems in Computer Graphics, this course looks at Poisson problems through the lens of 13 technical papers selected by the lecturers. The lecturers will highlight trends, similarities, differences, and historical threads through the papers.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Full Conference Supporter
Full Conference
DescriptionUltra-high-resolution sensors capture fine details critical for visual perception, but acquiring all pixels at full resolution is often infeasible under realistic constraints. We introduce a real-time, task-aware foveated imaging framework that learns a sensor attention policy to dynamically allocate pixel bandwidth, extensively validated in simulation and on a 200-megapixel prototype.
Technical Paper
Research & Education
Geometry
Hardware
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionWe present a method for computing error-bounded Hausdorff distance between triangle meshes on the GPU. By eliminating the serial bottleneck in traditional branch-and-bound algorithms, our approach achieves high-throughput performance and enables geometry processing applications that require guaranteed error control.
Technical Paper
Research & Education
Geometry
Modeling
Real-Time
Rendering
Full Conference Supporter
Full Conference
DescriptionPR-Cage is an automated framework for generating high-quality cages—simplified mesh proxies that tightly enclose complex geometry. Using a staircase relaxation of a thickness parameter and constrained QEM optimization, it achieves extreme simplification while preserving shape fidelity, enabling efficient contact simulation, deformation, and other graphics applications.
Talks
New Technologies
Production & Animation
Animation
Dynamics
Geometry
Industry Insight
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionThe talk presents fast, modular approximations of Curly‑Cue’s methods for kinky‑coily hair—switchbacks, phase locking, and period skipping—integrated into Netflix Animation Studio’s Houdini-based grooming pipeline, prioritizing artistic control, simplicity, and interactivity over strict physical accuracy while achieving natural-looking hairstyles.
Talks
Production & Animation
Research & Education
Animation
Artificial Intelligence/Machine Learning
Audio
Computer Vision
Performance
Full Conference Supporter
Full Conference
DescriptionFor Predator: Badlands, Wētā FX VFX Supervisor Sheldon Stopsack and Animation Supervisor Karl Rapley present how they developed Dek, a fully digital facial performance anchored in a practical on-set suit, to sustain a nuanced emotional arc without compromising the Predator’s franchise-defining silhouette and ferocity.
Technical Paper
Research & Education
Animation
Geometry
Hardware
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Immersive Pavilion
Arts & Design
Gaming & Interactive
New Technologies
Research & Education
Animation
Art
Artificial Intelligence/Machine Learning
Audio
Augmented Reality
Computer Vision
Digital Twins
Education
Ethics and Society
Fabrication
Games
Generative AI
Haptics
Hardware
Image Processing
Modeling
Performance
Pipeline Tools and Work
Real-Time
Rendering
Scientific Visualization
Simulation
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionPrism is an Apple Vision Pro application that renders intraoperative images as volumetric content in a spatial workspace. It combines dual rendering pipelines, touch-free voice and gesture control, 3D measurement tools, and multi-user collaboration via SharePlay, showing how spatial computing can improve imaging workflows in interventional radiology.
Technical Paper
Research & Education
Geometry
Modeling
Rendering
Simulation
Full Conference Supporter
Full Conference
DescriptionA Walk on Spheres-based Monte Carlo PDE solver that reuses full random walk paths via fixed probes, achieving lower variance and faster convergence on mixed-boundary Laplace, Poisson, and screened Poisson problems.
Talks
Production & Animation
Animation
Pipeline Tools and Work
Real-Time
Full Conference Supporter
Full Conference
DescriptionWalt Disney Animation Studios Technology developed a new automated data management system that migrates production resources between storage tiers and cleans up unneeded data throughout a film’s lifetime. The system ensures no disruption to end users and maintains data integrity while also aiming to be user-friendly, scalable, and observable.
Talks
Arts & Design
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Animation
Artificial Intelligence/Machine Learning
Augmented Reality
Capture/Scanning
Computer Vision
Digital Twins
Games
Geometry
Hardware
Image Processing
Performance
Pipeline Tools and Work
Rendering
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionOur system solves difficult VFX matchmoves by fusing LiDAR geometry with Deep Optical Flow. Unlike traditional corner-detection, it leverages dense geometric constraints to handle motion blur and dynamic occlusion. The system features automated actor masking and a novel re-anchoring architecture that eliminates drift, delivering accurate camera solves for production pipelines.
Technical Paper
Research & Education
Animation
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionWe extend Progressive Dynamics [Zhang et al. 2024, 2025] from cloth and shells to volumetric finite elements, enabling an efficient level-of-detail (LOD) animation-design pipeline with predictive coarse-resolution previews for rapid iteration toward a final high-resolution volumetric elastodynamics animation.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Fabrication
Generative AI
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionWe introduce a method for simplifying photorealistic scenes by progressively removing elements based on semantic importance. Our framework systematically removes elements in a hierarchical order, using generative inpainting to maintain photorealism throughout the process. This generates plausible photograph sequences, enabling applications like content-aware decluttering, layer decomposition, and interactive editing.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Full Conference Supporter
Full Conference
DescriptionProx-E is a training-free framework for fine-grained 3D shape editing that uses a primitive-based geometric proxy to explicitly control structural changes. By editing this proxy with a VLM and guiding a 3D generative model accordingly, it enables localized, identity-preserving edits that outperform prior 2D-driven and training-based 3D editing approaches.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Modeling
Real-Time
Rendering
Full Conference Supporter
Full Conference
DescriptionPureSample is a neural BRDF representation that learns material appearance directly from forward random walks on microgeometry. It enables efficient BRDF evaluation, importance sampling, and pdf evaluation using a flow-based sampler and a lightweight view-dependent albedo network, supporting complex homogeneous and spatially varying materials.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Fabrication
Geometry
Modeling
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Augmented Reality
Rendering
Simulation
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionWe built an ultrarealistic display system capable of reproducing 3D scenes with such fidelity that they are indistinguishable from their real counterparts. Then, we used the display to measure how reducing resolution and contrast affects perceived realism.
Technical Paper
Research & Education
Geometry
Modeling
Full Conference Supporter
Full Conference
DescriptionWe introduce an implicit method for medial axis extraction from point clouds and meshes, addressing challenges in noisy or incomplete data. By connecting the signed distance field and the medial field, we reformulate the problem as implicit reconstruction, offering a novel, robust solution for accurate medial axis computation.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionWe present R-DMesh, a novel feed-forward video-guided mesh animation framework that solves the pose misalignment dilemma between an initial mesh and a condition video through jump offset modeling. It generates high-fidelity, motion aligned 4D animations and enables downstream applications like pose retargeting and holistic 4D generation.
Technical Paper
Research & Education
Generative AI
Geometry
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionGaussian Splatting enables high-quality, real-time novel-view synthesis, but extracting detailed 3D geometry remains challenging. We introduce a rasterized approach to render the depth and normal maps for general 3D Gaussian primitives. Our method significantly enhances shape reconstruction accuracy while maintaining the computational efficiency of Gaussian Splatting.
Technical Paper
Research & Education
Lighting
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionWe introduce a spatially blended radiance-caching framework for differentiable path tracing that improves speed and robustness in material recovery under unknown lighting.
By jointly optimizing cache, materials, and blending, the method produces physically meaningful materials that generalize to new lighting conditions.
By jointly optimizing cache, materials, and blending, the method produces physically meaningful materials that generalize to new lighting conditions.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionNeural radiance fields and Gaussian splats often produce artifacts in low light, high dynamic range, or with rapid motion. Instead, by training at the granularity of individual photons, as enabled by single-photon cameras, we can achieve sharp, reliable reconstructions even in these challenging conditions.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Modeling
Rendering
Simulation
Full Conference Supporter
Full Conference
DescriptionRaNAD is the first spatiotemporal neural denoiser for deep Monte Carlo renderings. It handles semi-structured deep images with variable bin counts per pixel through ragged neighborhood attention, and exploits temporal context across up to seven frames, greatly advancing in both quality and temporal stability for denoising deep images.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Modeling
Rendering
Simulation
Full Conference Supporter
Full Conference
DescriptionWe introduce a novel Rao-Blackwellization technique for variance reduction in Markov chain Monte Carlo light transport. We improve both Metropolis Light Transport and the current state-of-the-art, Jump Restore Light Transport, consistently outperforming traditional variance reduction, accelerating convergence, and establishing a new state-of-the-art in MCMC-based rendering under equal-time and equal-sample evaluations.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Generative AI
Geometry
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionRaster2Seq is an anchor-based autoregressive formulation that frames floorplan reconstruction as scalable, sequential polygon prediction, transforming rasterized floorplan images into vectorized format. Through using our proposed labeled polygon sequence representation, the method jointly encodes spatial and semantic attributes, overcoming fixed-query budget limitations while significantly improving robustness as floorplan complexity increases.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Generative AI
Geometry
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionRCGP introduces a type system for graphics programming that statically enforces resource contracts, the agreements on type, layout, binding, and synchronization between shaders, pipelines, and host code. Through contracts, modules, combinators, and witnesses, RCGP catches descriptor mismatches, layout drift, and missing barriers at compile time, eliminating common classes of errors.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Robotics
Simulation
Full Conference Supporter
Full Conference
DescriptionWe present a reinforcement learning-based retargeting method that transfers human motion to diverse morphologies, including humanoids and a quadruped, without artifacts such as foot-sliding, hovering, or self-penetration. A bilevel optimization framework refines upper-loop retargeting parameters defined via sparse semantic correspondences while training the lower-loop RL policy, eliminating manual correspondence tuning.
Technical Paper
Research & Education
Geometry
Hardware
Modeling
Real-Time
Rendering
Full Conference Supporter
Full Conference
DescriptionWe introduce a GPU-based algorithm for real-time signed distance field generation from quadratic vector graphics. By treating SDF computation as parallel stroke rendering, the method achieves substantial speed improvements over CPU approaches, enabling dynamic text, interactive user interfaces, and real-time distance-based visual effects.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Real-Time
Rendering
Full Conference Supporter
Full Conference
DescriptionWe propose a method to overcome a limitation in ReSTIR path tracing by introducing a surface point mapping to reuse samples across frames containing meshes with different topologies and enabling real-time LoD rendering with ReSTIR.
Technical Paper
Research & Education
Games
Real-Time
Rendering
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Augmented Reality
Rendering
Simulation
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionThis paper investigates how avatar and face representation influence the perceptual evaluation of synthesized co-speech gestures. Across three controlled experiments, it shows that visual presentation systematically biases judgments of motion quality and provides practical recommendations for fairer, more reliable benchmarking of gesture generation systems.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Games
Generative AI
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionWave-coding methods for sound propagation in games are very memory-intensive. We present Reciprocal Latent Fields leveraging Riemannian metric learning to reduce memory usage by orders of magnitude. Our method infers relevant acoustic parameters nearly instantaneously and results in perceptually identical sound quality to the ground truth wave coding.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Generative AI
Geometry
Modeling
Rendering
Simulation
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionRef-DGS is a dual Gaussian splatting framework that decouples surface reconstruction and specular reflection, modeling near- and far-field specular reflections without ray tracing, achieving efficient, high-quality surface reconstruction and novel view synthesis.
Spatial Storytelling
New Technologies
Performance
Rendering
Spatial Computing
Full Conference Supporter
Full Conference
Experience
DescriptionREfract is a hybrid performance where a dancer and avatars coexist, evolve, and shift agency
through motion capture and live rendering. In a 15-minute performance, set in a minimalist
desert landscape with floating letters, the piece explores transformation, spatial perception, movement-driven storytelling, presence, autonomy, physical-digital interaction.
through motion capture and live rendering. In a 15-minute performance, set in a minimalist
desert landscape with floating letters, the piece explores transformation, spatial perception, movement-driven storytelling, presence, autonomy, physical-digital interaction.
Immersive Pavilion
Arts & Design
Gaming & Interactive
New Technologies
Research & Education
Animation
Art
Artificial Intelligence/Machine Learning
Audio
Augmented Reality
Computer Vision
Digital Twins
Education
Ethics and Society
Fabrication
Games
Generative AI
Haptics
Hardware
Image Processing
Modeling
Performance
Pipeline Tools and Work
Real-Time
Scientific Visualization
Simulation
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionReframe is a multiplayer VR experience where groups collectively edit 360° photographs using generative AI. Participants inhabit panoramic environments then point at areas while speaking about what they want to see. The system transcribes their dialogue into prompts and composites AI-generated imagery into the shared scene to elicit further prompting.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Lighting
Modeling
Real-Time
Rendering
Full Conference Supporter
Full Conference
DescriptionRelit-LiVE produces physically consistent and temporally stable relighting results without requiring prior knowledge of the camera pose. It utilizes raw reference images to ensure realistic lighting effects and introduces a joint prediction framework generating relighting videos with frame-level warped environment maps, consistently outperforming state-of-the-art methods across benchmarks.
Course
Full Conference Supporter
Full Conference
DescriptionLearn to build automated character generation pipelines using ComfyUI's REST/WebSocket APIs. Deploy state-of-the-art open source image generation and 3d generation models (Qwen, Hunyuan3D) via Docker and cloud GPU providers, then build Typer CLI tools that transform text prompts into rigged 3D characters through programmatic workflow execution: generate→turnaround→mesh→rig.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Generative AI
Full Conference Supporter
Full Conference
DescriptionReRoPE is a plug-and-play framework for controllable video generation that achieves precise camera control by injecting relative pose information into the underutilized low-frequency bands of standard Rotary Positional Embeddings (RoPE).
Production Session
Arts & Design
New Technologies
Production & Animation
Research & Education
Art
Artificial Intelligence/Machine Learning
Display
Education
Generative AI
Image Processing
Industry Insight
Performance
Pipeline Tools and Work
Rendering
Full Conference Supporter
Full Conference
DescriptionAdapting the 1939 classic The Wizard of Oz for the 16K x 16K immersive canvas of the Las Vegas Sphere presented a unique technical paradox: how to translate nearly century-old, 4:3 analog footage into a fully immersive environment. Initiated at a time when viable generative video models did not yet exist, and requiring bespoke solutions beyond the capabilities of emerging commercial tools, this session reveals the collaborative R&D workflow between Magnopus and Google DeepMind.
Panelists will dissect the hybrid production pipeline used to solve the project's most complex challenges: Generative Super Resolution, Character Outpainting, and AI Performance Generation.
First, the panel will address the image fidelity challenge. On the Sphere, the original film grain appeared as massive visual noise, and the limited resolution resulted in indistinct, "blobby" shapes. The team will detail the development of novel upscaling architectures designed to fundamentally modernize the footage. We will demonstrate how these custom models suppressed legacy artifacts and synthesized new high-frequency detail, creating sharpness and texture comparable to modern digital cinema, while maintaining the temporal coherence required for large-format playback.
Second, the session explores the integration of generative elements into a traditional VFX environment. While traditional vendor workflows handled the massive environmental extensions, the team utilized Generative AI to solve the complex character continuity challenges. We will detail the use of Character Outpainting to seamlessly complete characters partially cropped by the original 4:3 frame, alongside the generation of new AI Character Performances for those who walked completely out of view. The team will discuss the strategies used to ensure these generated assets matched the distinct physical language of the original actors.
Finally, the session covers the engineering hurdles of orchestrating this hybrid pipeline. We will examine how the team managed the ingest of disparate high-resolution generative elements to deliver a unified, temporally stable 16K master for the world’s largest LED screen.
Since its debut, this pioneering effort has become one of the most commercially successful generative AI deployments in entertainment history, driving record-breaking audiences at Sphere. Validating the success of this hybrid R&D approach, the project earned the prestigious PGA Innovation Award and the VES Award for Outstanding Visual Effects in a Special Venue Project. Attendees will leave with a proven roadmap for navigating the intersection of generative AI research, traditional VFX, and next-generation immersive storytelling
Panelists will dissect the hybrid production pipeline used to solve the project's most complex challenges: Generative Super Resolution, Character Outpainting, and AI Performance Generation.
First, the panel will address the image fidelity challenge. On the Sphere, the original film grain appeared as massive visual noise, and the limited resolution resulted in indistinct, "blobby" shapes. The team will detail the development of novel upscaling architectures designed to fundamentally modernize the footage. We will demonstrate how these custom models suppressed legacy artifacts and synthesized new high-frequency detail, creating sharpness and texture comparable to modern digital cinema, while maintaining the temporal coherence required for large-format playback.
Second, the session explores the integration of generative elements into a traditional VFX environment. While traditional vendor workflows handled the massive environmental extensions, the team utilized Generative AI to solve the complex character continuity challenges. We will detail the use of Character Outpainting to seamlessly complete characters partially cropped by the original 4:3 frame, alongside the generation of new AI Character Performances for those who walked completely out of view. The team will discuss the strategies used to ensure these generated assets matched the distinct physical language of the original actors.
Finally, the session covers the engineering hurdles of orchestrating this hybrid pipeline. We will examine how the team managed the ingest of disparate high-resolution generative elements to deliver a unified, temporally stable 16K master for the world’s largest LED screen.
Since its debut, this pioneering effort has become one of the most commercially successful generative AI deployments in entertainment history, driving record-breaking audiences at Sphere. Validating the success of this hybrid R&D approach, the project earned the prestigious PGA Innovation Award and the VES Award for Outstanding Visual Effects in a Special Venue Project. Attendees will leave with a proven roadmap for navigating the intersection of generative AI research, traditional VFX, and next-generation immersive storytelling
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Generative AI
Geometry
Modeling
Rendering
Simulation
Full Conference Supporter
Full Conference
DescriptionWe propose a variance-adaptive voxel grid that adjusts size based on local SDF variance. Using a flat spatial hash table, it achieves constant-time access, memory efficiency, and full GPU parallelism. Additionally, a parallel quad-tree structure supports GPU-accelerated Gaussian Splatting rendering, enabling effective control over splat density.
Art Paper
Arts & Design
Research & Education
Animation
Art
Augmented Reality
Dynamics
Hardware
Lighting
Performance
Real-Time
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionResonance transforms a traditionally solitary meditative practice into a shared spatial experience by externalizing a meditator's neural rhythms as architectural-scale light and motion. Rather than treating neurofeedback as private data, the work situates cognition within an immersive environment that can be collectively sensed and inhabited. Bridging contemplative traditions with computational design, it expands bio-responsive systems beyond subjective interaction toward shared, embodied engagement and collective awareness.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Real-Time
Rendering
Full Conference Supporter
Full Conference
DescriptionReSTIR accelerates path tracing by reusing samples between pixels. However, existing methods are limited by the sampling quality of path tracing, making them inefficient for scenes with caustics and hard-to-reach lights. We develop a ReSTIR variant incorporating bidirectional path tracing that significantly improves the sampling quality in these scenes.
Emerging Technologies
Arts & Design
Gaming & Interactive
New Technologies
Research & Education
Art
Artificial Intelligence/Machine Learning
Audio
Augmented Reality
Capture/Scanning
Computer Vision
Digital Twins
Display
Dynamics
Education
Fabrication
Games
Generative AI
Graphics Systems Architecture
Haptics
Hardware
Image Processing
Performance
Physical AI
Pipeline Tools and Work
Real-Time
Robotics
Scientific Visualization
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionWe propose an integration of render and display that reduces interactive latency far below delays in current systems -- our measurements show click-to-photon latency below 2 ms. We will demonstrate a low-resolution hardware prototype of this integrated pipeline along with a high-resolution interactive simulator of alternative display scans.
Technical Paper
Research & Education
Animation
Fabrication
Haptics
Simulation
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionWe present a computational fabrication pipeline for retrofitting 3D objects with real-time multi-touch sensing without modifying their interiors. From a 3D scan, our method routes and optimizes surface-conforming electrode layouts under geometric, fabrication, and hardware constraints. We guide assembly via projection to enable precise spatial input sensing on curved objects.
Computer Animation Festival
DescriptionNo one likes to be a third wheel on a first date. Right Hand
Man pushes the boundaries of this concept with its two main
characters, Lucas and Cade, twin brothers who are a human
and a hand puppet, respectively. We follow their journey as
Lucas tries to navigate his first date with the lovely Cristine at a
fine dining restaurant, attempting to hide Cade away to
masquerade as an average guy. All Cade wants to do, though,
is get in on the excitement and romance of the experience. Will
Cade make himself known to Cristine? Will Lucas learn that
hiding your true self is the worst foundation for a first date?
Watch and find out!
Man pushes the boundaries of this concept with its two main
characters, Lucas and Cade, twin brothers who are a human
and a hand puppet, respectively. We follow their journey as
Lucas tries to navigate his first date with the lovely Cristine at a
fine dining restaurant, attempting to hide Cade away to
masquerade as an average guy. All Cade wants to do, though,
is get in on the excitement and romance of the experience. Will
Cade make himself known to Cristine? Will Lucas learn that
hiding your true self is the worst foundation for a first date?
Watch and find out!
Talks
Production & Animation
Research & Education
Animation
Artificial Intelligence/Machine Learning
Audio
Computer Vision
Performance
Full Conference Supporter
Full Conference
DescriptionWe present Rigs2Emotion, a ML tool based on rig controllers values to classify emotions of animations frames accurately. It can be used on any production, is character- and rig-specific and is much more data-efficient than more usual method using audio or image fine-tuned pre-trained models.
Appy Hour
Arts & Design
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Animation
Art
Artificial Intelligence/Machine Learning
Audio
Augmented Reality
Computer Vision
Digital Twins
Education
Ethics and Society
Games
Generative AI
Geometry
Image Processing
Math Foundations and Theory
Modeling
Performance
Physical AI
Pipeline Tools and Work
Real-Time
Robotics
Scientific Visualization
Simulation
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionMelody Slot Machine has dials with the staves, which can be rotated to change melody variations which are generated by AI-based melody-morphing method. The microcomputer on RoboSax receives MIDI notes from Melody Slot Machine and move servomotors so that the fingering corresponds to the note number.
Technical Paper
Research & Education
Animation
Geometry
Hardware
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionWe introduce a barrier-free optimization framework for nonpenetration elastodynamic simulation using a smooth augmented Lagrangian formulation and efficient active-set exploration. Unlike IPC barriers, our method avoids TOI locking, maintains a compact anticipatory constraint set, and delivers faster, robust, visually faithful penetration-free simulation for demanding contact-rich applications.
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Fabrication
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionA mesh-free approach for computing bounded biharmonic skinning weights. Our method works well even in presence of severe mesh quality degradation and can incorporate artistic control via weight-painting. It relies on a differentiable field and incorporates boundary handling via hardware accelerated ray-tracing.
Technical Paper
Research & Education
Lighting
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionWe present a simple, robust, and consistent estimator for boundary path integrals in physics-based differentiable rendering. By reformulating boundary integration with kernel-density estimation, our method avoids fragile measure-zero sampling and costly reparameterization. It matches finite-difference derivatives and outperforms state-of-the-art baselines in synthetic inverse-rendering benchmarks with improved efficiency and lower variance.
Technical Paper
Research & Education
Fabrication
Geometry
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionWe introduce an accurate and precise generalized winding number method for trimmed NURBS patches based on integrating along boundary curves. For points near a patch, we add a correction term determined via line-surface intersections. The method enables robust inside-outside containment queries that respect curved features in complex CAD models.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Modeling
Real-Time
Rendering
Full Conference Supporter
Full Conference
DescriptionWe present a practical method for optimizing textures directly inside production rendering engines from inconsistent AI-generated images. By handling geometric mismatch and unreliable details, our approach avoids blur and ghosting, enabling robust texture refinement without requiring differentiable rendering or changes to existing pipelines.
Technical Paper
Research & Education
Geometry
Hardware
Modeling
Real-Time
Rendering
Full Conference Supporter
Full Conference
DescriptionVectorizing 3D scenes into 2D vector images requires planar maps to define solid non-overlapping regions. Existing planar map methods are slow and susceptible to numerical instability. We introduce a robust and efficient method for constructing planar maps, using a spatial hierarchy as the fundamental representation of the planar map.
Computer Animation Festival
Games Summit
Gaming & Interactive
Games
Full Conference Supporter
Full Conference
DescriptionRocketship Mai follows a fearless six-year-old girl who is on a daring mission
to tackle Time (literally) so she can spend more moments with her ailing father.
With her anxious copilot Bunny by her side, Mai speeds through the galaxy to box
clocks, battle oversized foes (like... a million times larger), and take on anything the
cosmos throws in her way.
As the fight escalates, however, Mai discovers that time can’t be stopped — and what
remains is something far more freeing: learning to be fully present with the person she
loves most.
to tackle Time (literally) so she can spend more moments with her ailing father.
With her anxious copilot Bunny by her side, Mai speeds through the galaxy to box
clocks, battle oversized foes (like... a million times larger), and take on anything the
cosmos throws in her way.
As the fight escalates, however, Mai discovers that time can’t be stopped — and what
remains is something far more freeing: learning to be fully present with the person she
loves most.
Technical Paper
Research & Education
Animation
Fabrication
Haptics
Simulation
Virtual Reality
Full Conference Supporter
Full Conference
DescriptionWe present a role-aware virtual agent navigation system that generates consistent, role-aligned behaviors. Using multimodal large language models (MLLMs), our approach interprets scene context, user state, and role-context instructions to produce navigation decisions and stylized paths, enabling agents to respond dynamically and behave coherently in immersive environments such as AR.
Spatial Storytelling
Gaming & Interactive
Games
Performance
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionRoll for Reality is a live, improvised Dungeons & Dragons show integrating tabletop roleplaying, motion capture, AI-assisted narrative tools, and real-time XR visualization. The project explores hybrid XR storytelling frameworks that connect improvisational performance with live virtual production environments.
Educator's Forum
Arts & Design
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Animation
Art
Computer Vision
Digital Twins
Education
Games
Modeling
Pipeline Tools and Work
Rendering
Simulation
Spatial Computing
Full Conference Supporter
Full Conference
Experience
DescriptionThis paper presents an introductory virtual production (VP) course built around an LED-wall stage, where students learn VP through embodied, pipeline-centered practice. By integrating cinematography, real-time rendering, tracking, and collaborative workflows, the course develops ecosystem fluency and workflow literacy across film, animation, and game design disciplines.
Talks
Gaming & Interactive
New Technologies
Production & Animation
Research & Education
Animation
Artificial Intelligence/Machine Learning
Computer Vision
Education
Games
Generative AI
Image Processing
Simulation
Full Conference Supporter
Full Conference
DescriptionRoto-Lite is a pixels-to-sequence transformer model that parses regions of portrait captures and outputs an efficient set of spline shapes. Our aim is to offer minimal friction between the outputs of our model and its adoption by artists, as compared to more traditional raster-based segmentation models.
Computer Animation Festival
DescriptionIn an abstract and symbolic world where gravity is reversed, a unique bond forms between a grandfather and his grandson. As Saba, the grandfather guides the boy through this surreal reality, a violent storm strikes, tearing one of the ropes anchoring their home - and with it, their life together.
After suffering a devastating loss of Saba, the boy embarks on an emotional journey in search of closure. Through a stylized and poetic visual language, the film explores life’s fundamental questions: the paths we take, separation from those we love, and the endless effort to find closure, reconcile loss, and move forward.
After suffering a devastating loss of Saba, the boy embarks on an emotional journey in search of closure. Through a stylized and poetic visual language, the film explores life’s fundamental questions: the paths we take, separation from those we love, and the endless effort to find closure, reconcile loss, and move forward.
Computer Animation Festival
Games Summit
Gaming & Interactive
Games
Full Conference Supporter
Full Conference
DescriptionSalvation plunges viewers into a world on the brink of chaos, where champions confront both internal and external conflicts that threaten the balance of their realm. The cinematic follows key heroes as they face moral dilemmas, epic battles, and alliances that define the fate of their world. Through cinematic storytelling, dynamic action, and striking visual design, the trailer conveys the tension, stakes, and grandeur of the League of Legends universe.
Technical Paper
Research & Education
Lighting
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionWe propose sample matching, a variance reduction principle for differentiable volume rendering. By jointly estimating scattering and transmittance gradient components at shared sample positions, our method reduces gradient variance by up to 80%, significantly improving convergence speed and reconstruction quality in volumetric inverse rendering.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Full Conference Supporter
Full Conference
DescriptionSAND is an efficient neural implicit surface representation framework that accelerates sampling by adapting network depth to spatial importance and local geometric complexity. By combining a volumetric depth map with a tailed MLP for adaptive termination, it significantly improves inference-time query efficiency while preserving high-fidelity reconstruction and enabling neural level-of-detail.
Technical Paper
Research & Education
Geometry
Hardware
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionWe present a GPU algorithm for constructing large-scale 3D Voronoi and power diagrams. By combining directional geometric culling with hierarchical best-first traversal, our method efficiently handles tens of millions of points across diverse spatial distributions. Applied to mesh-based neural rendering, it removes a key computational bottleneck, enabling higher-fidelity scene reconstructions.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Full Conference Supporter
Full Conference
DescriptionWe introduce a spatially discrete formulation of the Schrödinger bridge problem on meshes and grids that enables structure-preserving interpolation between probability distributions. We demonstrate our approach across mesh- and grid-based applications, including displacement interpolation, shape interpolation, and color histogram manipulation, highlighting its ability to achieve geometric fidelity with computational efficiency.
Emerging Technologies
Arts & Design
Gaming & Interactive
New Technologies
Research & Education
Art
Artificial Intelligence/Machine Learning
Audio
Augmented Reality
Capture/Scanning
Computer Vision
Digital Twins
Display
Dynamics
Education
Fabrication
Games
Generative AI
Graphics Systems Architecture
Haptics
Hardware
Image Processing
Performance
Physical AI
Pipeline Tools and Work
Real-Time
Robotics
Scientific Visualization
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionSearchable.City is an open-vocabulary semantic atlas, built by running a vision-language model (VLM) on millions of street-view images to create a searchable visual index of the city. It turns street-level imagery into a map of visual patterns, revealing where architectural styles, infrastructure, culture, and objects concentrate.
Talks
Production & Animation
Animation
Modeling
Pipeline Tools and Work
Simulation
Full Conference Supporter
Full Conference
DescriptionThis talk presents a hybrid workflow of procedural systems informing physics-driven simulations used to craft the natural crowds in Disney and Pixar’s Hoppers. We will present details of crowds workflows using both SideFx Houdini, and Pixar’s proprietary Presto PCF toolsets to produce hordes of snakes, butterflies, seagulls, and more!
Technical Paper
Research & Education
Animation
Artificial Intelligence/Machine Learning
Generative AI
Geometry
Modeling
Simulation
Full Conference Supporter
Full Conference
DescriptionWe introduce a framework that automates the transformation of static anime illustrations into manipulatable 2.5D models. Our approach decomposes a single image into fully inpainted, semantically distinct layers with inferred drawing orders — up to 19 layers including hair, face, eyes, clothing, accessories, and more.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Rendering
Robotics
Full Conference Supporter
Full Conference
DescriptionSegviGen is a framework for 3D part segmentation which unifies three settings in one architecture: interactive part segmentation, full segmentation, and 2D segmentation map–guided full segmentation. It improves over sota by 40% on interactive segmentation and by 15% on full segmentation, while using only 0.32% of the training data.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Fabrication
Generative AI
Modeling
Rendering
Full Conference Supporter
Full Conference
DescriptionWe introduce a generative framework that transforms traditional statistical graphics into memorable, engaging pictorial charts. It utilizes structural alignment to anchor spatial layouts and semantic alignment to transfer expressive textures. This ensures the resulting visuals are artistically compelling and structurally consistent, advancing data-driven visual storytelling.
Immersive Pavilion
Arts & Design
Gaming & Interactive
New Technologies
Research & Education
Animation
Art
Artificial Intelligence/Machine Learning
Audio
Augmented Reality
Computer Vision
Digital Twins
Education
Ethics and Society
Fabrication
Games
Generative AI
Haptics
Hardware
Image Processing
Modeling
Performance
Pipeline Tools and Work
Real-Time
Scientific Visualization
Simulation
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionSentiSing is an affect-driven VR singing experience where vocal and musical affect jointly control generative visual changes in a forest environment, allowing singers to perceive and explore emotional expression through real-time immersive feedback.
Educator's Forum
Gaming & Interactive
Research & Education
Education
Games
Image Processing
Modeling
Real-Time
Full Conference Supporter
Full Conference
Experience
DescriptionThe Sew and Code assignment is designed to fuse programming, math, algorithms, art, and design and provide motivation to Computer Graphics students learning how to construct Bezier curves using the DeCastlejau Algorithm. Students cycle between sewing then programming steps of the algorithm, building up String Art and interactive programs simultaneously.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Lighting
Modeling
Real-Time
Rendering
Full Conference Supporter
Full Conference
DescriptionAfter the summary presentations, attendees will participate in an interactive discussion. Outside the room will be a series of poster boards for authors to gather around with the audience. Authors are invited to bring any material related to their paper that could instigate further conversation such as printouts, posters, demos, or other presentation aids. The interactive discussions provide attendees more interaction with the individual authors.
Emerging Technologies
Arts & Design
Gaming & Interactive
New Technologies
Research & Education
Art
Artificial Intelligence/Machine Learning
Audio
Augmented Reality
Capture/Scanning
Computer Vision
Digital Twins
Display
Dynamics
Education
Fabrication
Games
Generative AI
Graphics Systems Architecture
Haptics
Hardware
Image Processing
Performance
Physical AI
Pipeline Tools and Work
Real-Time
Robotics
Scientific Visualization
Spatial Computing
Virtual Reality
Full Conference Supporter
Full Conference
Experience
DescriptionWe propose an agent grounded in the Free Energy Principle that negotiates intentions through physical interaction. Using a predictive coding-based RNN, the agent dynamically updates internal intentions and time constants to minimize prediction errors, enabling seamless synchronization with human choreography and tempo.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Full Conference Supporter
Full Conference
DescriptionShapeUp introduces a scalable framework for high-fidelity 3D asset editing. By casting editing as an image-conditioned, latent-to-latent translation, it enables precise local and global transformations while maintaining native 3D consistency. Our approach ensures seamless identity preservation without requiring manual spatial masks.
Technical Paper
Research & Education
Artificial Intelligence/Machine Learning
Geometry
Modeling
Real-Time
Rendering
Full Conference Supporter
Full Conference