BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Los_Angeles
X-LIC-LOCATION:America/Los_Angeles
BEGIN:DAYLIGHT
TZOFFSETFROM:-0800
TZOFFSETTO:-0700
TZNAME:PDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0700
TZOFFSETTO:-0800
TZNAME:PST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260805T205455Z
LOCATION:Room 408 B
DTSTART;TZID=America/Los_Angeles:20260721T104500
DTEND;TZID=America/Los_Angeles:20260721T121500
UID:siggraph_SIGGRAPH 2026_sess144@linklings.com
SUMMARY:Dynamic Scenes
DESCRIPTION:MTPano: Multi-Task Panoramic Scene Understanding via Label-Fre
 e Integration of Dense Prediction Priors\n\nMTPano is a robust multi-task 
 foundation model for panoramic scene understanding. Trained via a label-fr
 ee pipeline using perspective dense priors, it adopts a dual-stream archit
 ecture that resolves feature conflicts between depth, segmentation, and su
 rface normal estimations, achieving state-of-the...\n\n\nJingdong Zhang (T
 exas A&M University); Xiaohang Zhan and Lingzhi Zhang (Adobe); Yizhou Wang
  (Northeastern University, Adobe); and Zhengming Yu, Jionghao Wang, Wenpin
 g Wang, and Xin Li (Texas A&M University)\n---------------------\nSmoothMo
 tionVectors: Optimizing Your Content for Video Codecs in Free View Video C
 ompression\n\nConventional video codecs exploit spatial and temporal self-
 similarity to compress structured signals. We bring this principle to dyna
 mic scene reconstruction with 3D Gaussian Splatting by organizing deformat
 ion outputs through a tailored pipeline. The resulting motion vectors are 
 smoother across sp...\n\n\nMingyang Song (Disney Research Studios, ETH Zür
 ich); Yang Zhang (Disney Research Studios); Siyu Tang (ETH Zürich); and Tu
 nc Ozan Aydin (Disney Research Studios)\n---------------------\nCAGS: Colo
 r-Adaptive Volumetric Video Streaming with Dynamic 3D Gaussian Splatting\n
 \nCAGS is a novel color-adaptive system for 3D Gaussian Splatting that ena
 bles high-quality, immersive volumetric video streaming. By utilizing vect
 or quantization and reference images for color restoration, CAGS overcomes
  bandwidth limitations, delivering significantly faster speeds and superio
 r visu...\n\n\nDaheng Yin (Simon Fraser University); Yili Jin (McGill Univ
 ersity, Simon Fraser University); Jianxin Shi (Nankai University, Simon Fr
 aser University); Isaac Ding and Miao Zhang (Simon Fraser University); Fan
 gxin Wang (The Chinese University of Hong Kong); Zhaowu Huang (Fuzhou Univ
 ersity, Southeast University); Cong Zhang and Jiangchuan Liu (Simon Fraser
  University); and Fang Dong (Southeast University)\n---------------------\
 nTSMC: Time-varying 4D Scene Mesh Compression\n\nTSMC is a compression fra
 mework for time-varying scene meshes that separates static and dynamic reg
 ions, leverages mesh-enclosed volumes to create reference meshes for motio
 n prediction, and efficiently compresses dynamic displacements, enabling b
 andwidth-efficient 4D scene compression and real-time...\n\n\nGuodong Chen
  (Northeastern University), Libor Váša (University of West Bohemia), Amrit
 a Mazumdar (NVIDIA), and Mallesham Dasari (Northeastern University)\n-----
 ----------------\nSparse-to-Complete: From Sparse Image Captures to Comple
 te 3D Scenes\n\nWe introduce S2C-3D,  a novel sparse-view 3D reconstructio
 n framework for high-fidelity, complete scene reconstruction. It features 
 three components: a specialized diffusion model for scene-specific image r
 estoration, a training-free view-consistency conditioned sampling process 
 in the diffusion mod...\n\n\nYiyang Shen (State Key Lab of CAD and CG, Zhe
 jiang University); Yin Yang (The University of Utah); and Kun Zhou and Tia
 njia Shao (State Key Lab of CAD and CG, Zhejiang University)\n------------
 ---------\nArticulate That Object Part (ATOP): 3D Part Articulation from T
 ext and via Motion Personalization\n\nWe present a few-shot method for art
 iculating static 3D objects using text-guided motion personalization. Leve
 raging 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 ...\n\n\nAditya Vora, Sauradip N
 ag, Kai Wang, and Hao (Richard) Zhang (Simon Fraser University) and Aditya
  Vora\n---------------------\nDynamic Scenes - Interactive Discussion\n\nA
 fter the summary presentations, attendees will participate in an interacti
 ve discussion. Outside the room will be a series of poster boards for auth
 ors to gather around with the audience. Authors are invited to bring any m
 aterial related to their paper that could instigate further conversation s
 uch...\n\n\nInterest Area: Research & Education\n\nKeyword: Artificial Int
 elligence/Machine Learning, Geometry, Modeling, Rendering, Virtual Reality
 \n\nRegistration Category: Full Conference Supporter, Full Conference\n\nS
 ession Chair: Yi Zhou (Roblox)
END:VEVENT
END:VCALENDAR