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:20260805T205500Z
LOCATION:Room 403  B
DTSTART;TZID=America/Los_Angeles:20260722T104500
DTEND;TZID=America/Los_Angeles:20260722T123500
UID:siggraph_SIGGRAPH 2026_sess124@linklings.com
SUMMARY:Reconstruction & Sampling
DESCRIPTION:Mix3R: Mixing Feed-forward Reconstruction and Generative 3D Pr
 iors for Joint Multi-view Aligned 3D Reconstruction and Pose Estimation\n\
 nWe present Mix3R, a method that mixes feed-forward reconstruction and gen
 erative 3D priors using a Mixture-of-Transformer architecture for multi-vi
 ew aligned generative 3D reconstruction from sparse views.\n\n\nSiyou Lin 
 and Zhou Xue (Tsinghua University), Hongwen Zhang (Beijing Normal Universi
 ty), Liang An (Tsinghua University), Dongping Li and Shaohui Jiao (ByteDan
 ce Inc.), and Yebin Liu (Tsinghua University)\n---------------------\nDist
 ance Field Rasterization for End-to-End Mesh Reconstruction\n\nSDFRaster c
 ombines the efficiency of rasterization with signed distance fields for en
 d-to-end mesh reconstruction from multi-view images. By optimizing a tetra
 hedral SDF and extracting meshes during optimization, it produces accurate
 , complete, and compact surfaces without heuristic post-processing.\n\n\nJ
 inkai Cui, Kaiwen Song, Chumeng Niu, and Juyong Zhang (University of Scien
 ce and Technology of China)\n---------------------\nTopology Type Estimati
 on of Simulated 4D Image Data by Combining Downscaling and Convolutional N
 eural Networks\n\nThis project integrates a 4D camera model, synthetic 4D 
 datasets, and CNNs trained on downscaled data to analyse its 4D topology. 
 The results suggest that advanced 4D vision pipelines, particularly when p
 aired with significant computational resources, could enable scientific br
 eakthroughs unattainab...\n\n\nKhalil Hannouch and Stephan Chalup (The Uni
 versity of Newcastle) and Stephan Chalup\n---------------------\nNILE: Nes
 ted Interleaving of Low-Dimensional Elements\n\nWe introduce a novel modul
 ar modular meta-sampler architecture that bridges local subspace and globa
 l sampling, allowing integrator designers to employ specialized low-dimens
 ional samplers while still achieving high-dimensional uniformity.\n\n\nAbd
 alla Ahmed (Shenzhen University); Matt Pharr (NVIDIA); Victor Ostromoukhov
  (Université Claude Bernard Lyon 1, CNRS - LIRIS); and Hui Huang (Shenzhen
  University)\n---------------------\nGradient-based Design of Non-Uniform 
 Low-Discrepancy Samples\n\nThis paper introduces a first numerical optimiz
 ation method for approximately generating non-uniform low-discrepancy poin
 t sets with an arbitrary target distribution. It is based on a new differe
 ntiable estimator of non-uniform L2-discrepancy for a given point set.\n\n
 \nXiangyu Li (Brown University), Ege Ciklabakkal (University of Waterloo),
  Daniel Ritchie (Brown University), and Toshiya Hachisuka (University of W
 aterloo)\n---------------------\nBeyond Positional Encoding: A 5D Spatio-D
 irectional Hash Encoding\n\nThis paper introduces two new compact neural e
 ncodings, the hash-sphere and hash-grid-sphere, which represent all-freque
 ncy signals using hierarchical geodesic grids. While we demonstrated their
  effectiveness in neural path guiding, radiance field reconstruction, and 
 incident radiance caching, these...\n\n\nPhilippe Weier (Meta, Saarland Un
 iversity); Lukas Bode (Meta); Philipp Slusallek (Saarland University); and
  Adrián Jarabo and Sébastien Speierer (Meta)\n---------------------\nHi-SP
 AD: Video-Rate Hyperspectral Imaging and Inference with Single-Photon Came
 ras\n\nThis work develops a hyperspectral imager using a SPAD image sensor
  that achieves a spatial resolution of 1024 × 768 pixels across 123 spectr
 al bands spanning visible and near-infrared wavelengths, at a temporal rat
 e of 26 frames per second.\n\n\nHaejoon Lee (Carnegie Mellon University), 
 Mohit Gupta (University of Wisconsin-Madison), Vijayakumar Bhagavatula and
  Aswin Sankaranarayanan (Carnegie Mellon University), and Haejoon Lee\n---
 ------------------\nReconstruction & Sampling - Interactive Discussion\n\n
 After the summary presentations, attendees will participate in an interact
 ive discussion. Outside the room will be a series of poster boards for aut
 hors to gather around with the audience. Authors are invited to bring any 
 material related to their paper that could instigate further conversation 
 such...\n\n\nInterest Area: Research & Education\n\nKeyword: Artificial In
 telligence/Machine Learning, Geometry, Modeling, Rendering\n\nRegistration
  Category: Full Conference Supporter, Full Conference\n\nSession Chair: Sc
 ott Schaefer (Texas A&M University)
END:VEVENT
END:VCALENDAR