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:20260805T205512Z
LOCATION:Room 403 A
DTSTART;TZID=America/Los_Angeles:20260720T154500
DTEND;TZID=America/Los_Angeles:20260720T173500
UID:siggraph_SIGGRAPH 2026_sess110@linklings.com
SUMMARY:Humans & Hands - Reconstruction
DESCRIPTION:TwinPose: Person-Specific Subspaces for Multi-View 3D Pose Est
 imation\n\nTwinPose is an observation-driven framework for real-time multi
 -person 3D motion capture from sparse multi-view inputs in complex scenes.
  By constructing instance-aware "twin poses" that unify 2D pose semantics 
 and multi-view geometric consistency, it enables robust, efficient, and de
 tector-agnostic...\n\n\nWenwu Yang, Tianyi He, Jiwei Ding, Xun Wang, and R
 ong Zhang (Zhejiang Gongshang University) and Kun Zhou (Zhejiang Universit
 y)\n---------------------\nAMOR: Airborne Motion Reconstruction via Homoto
 py-Aware Trajectory Optimization\n\nExisting Human-Mesh-Recovery methods s
 truggle with dynamic airborne motions, often producing unstable and unreal
 istic results. We present a refinement approach that enforces physical con
 sistency by selecting reliable motion segments, applying homotopy-aware tr
 ajectory optimization. Experiments on in...\n\n\nChanha Kim and Jungdam Wo
 n (Seoul National University)\n---------------------\nFLASHand: Feed-forwa
 rd reLightable and Animatable Single-view Hand Reconstruction\n\nWe propos
 e FLASHand, the first feed-forward model to reconstruct high-fidelity, rel
 ightable, and animatable 3D hand avatars from a single RGB image. By lever
 aging NIMBLE priors and mesh-based 2D Gaussian Splatting, FLASHand achieve
 s instant, personalized reconstruction with disentangled appearance, ...\n
 \n\nLing-Xiao Zhang and Lin Gao (Institute of Computing Technology, Chines
 e Academy of Sciences; University of Chinese Academy of Sciences); Wei-Hon
 g He (South China University of Technology); Yu-Xuan Yang (East China Norm
 al University); Yunbing Xing (Institute of Computing Technology, Chinese A
 cademy of Sciences); Yu-Kun Lai (Cardiff University); and Yiqiang Chen (In
 stitute of Computing Technology, Chinese Academy of Sciences)\n-----------
 ----------\nAGILE: Hand-object Interaction Reconstruction from Video via A
 gentic Generation\n\nAGILE reconstructs 3D hand-object interactions from m
 onocular videos via agentic generation. It produces complete, simulation-r
 eady object meshes and robustly estimates object pose without SfM, enablin
 g accurate and stable results on challenging real-world data.\n\n\nJin-Chu
 an Shi, Binhong Ye, Tao Liu, Xiaoyang Liu, Yangjinhui Xu, Junzhe He, Zeju 
 Li, and Hao Chen (Zhejiang University) and Chunhua Shen (Zhejiang Universi
 ty, Zhejiang University of Technology)\n---------------------\nEgoForce: F
 orearm-Guided Camera-Space 3D Hand Pose from a Monocular Egocentric Camera
 \n\nEgoForce enables real-time 3D hand pose, shape, and absolute position 
 recovery from a single head-mounted RGB camera. Designed for lightweight s
 mart glasses, it uses forearm cues and camera-aware ray-space lifting to p
 rovide precise tracking across diverse camera models and device setups.\n\
 n\nChristen Millerdurai (DFKI, Max Planck Institute for Informatics); Shao
 xiang Wang and Yaxu Xie (DFKI); Vladislav Golyanik (Max Planck Institute f
 or Informatics); and Didier Stricker and Alain Pagani (DFKI)\n------------
 ---------\nOmniHands: Robust Motion Capture of Interactive Hands via A Ver
 satile Transformer\n\nOmniHands is a universal architecture for hand recon
 struction across diverse scenarios, supporting single-hand reconstruction,
  two-hand interaction reconstruction, temporal inputs, and multi-view inpu
 ts. It achieves this within a unified framework by first tokenizing the fe
 atures of each hand and th...\n\n\nDixuan Lin (Beijing Normal University, 
 Stanford University); Yuxiang Zhang and Mengcheng Li (Tsinghua University)
 ; Wei Jing, Qi Yan, and Qianying Wang (Lenovo Group Ltd); Yebin Liu (Tsing
 hua University); Hongwen Zhang (Beijing Normal University); and Dixuan Lin
 \n---------------------\nHumans & Hands - Reconstruction - Interactive Dis
 cussion\n\nAfter the summary presentations, attendees will participate in 
 an interactive discussion. Outside the room will be a series of poster boa
 rds for authors to gather around with the audience. Authors are invited to
  bring any material related to their paper that could instigate further co
 nversation such...\n\n\nInterest Area: Research & Education\n\nKeyword: An
 imation, Artificial Intelligence/Machine Learning, Generative AI, Geometry
 , Modeling, Simulation, Virtual Reality\n\nRegistration Category: Full Con
 ference Supporter, Full Conference\n\nSession Chair: Ruizhen Hu (Shenzhen 
 University (SZU))
END:VEVENT
END:VCALENDAR