BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Asia/Hong_Kong
X-LIC-LOCATION:Asia/Hong_Kong
BEGIN:STANDARD
TZOFFSETFROM:+0800
TZOFFSETTO:+0800
TZNAME:HKT
DTSTART:19911015T033000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20251218T030657Z
LOCATION:Meeting Room S221\, Level 2
DTSTART;TZID=Asia/Hong_Kong:20251218T110100
DTEND;TZID=Asia/Hong_Kong:20251218T111200
UID:siggraphasia_SIGGRAPH Asia 2025_sess153_papers_1066@linklings.com
SUMMARY:SRBTrack: Terrain-Adaptive Tracking of a Single-Rigid-Body Charact
 er Using Momentum-Mapped Space-Time Optimization
DESCRIPTION:Hanyang Cao (Hanyang University), Heyuan Yao and Libin Liu (Pe
 king University), and Taesoo Kwon (Hanyang University)\n\nGenerating reali
 stic and robust motion for virtual characters under complex physical condi
 tions, such as irregular terrain, real-time control scenarios, and externa
 l disturbances, remains a key challenge in computer graphics.\nWhile deep 
 reinforcement learning has enabled high-fidelity physics-based character a
 nimation, such methods often suffer from limited generalizability, as lear
 ned controllers tend to overfit to the environments they were trained in. 
 In contrast, simplified models, such as single rigid bodies, offer better 
 adaptability, but traditionally require hand-crafted heuristics and can on
 ly handle short motion segments. In this paper, we present a general learn
 ing framework that trains a single-rigid-body (SRB) character controller f
 rom long and unstructured datasets, without the reliance on human-crafted 
 rules. Our method enables zero-shot adaptation to diverse environments and
  unseen motion styles.  The resulting controller generates expressive and 
 physically plausible motions in real time and seamlessly integrates with h
 igh-level kinematic motion planners without retraining, enabling a wide ra
 nge of downstream tasks.\n\nRegistration Category: Full Access, Full Acces
 s Supporter\n\nSession Chair: Kai Wang (Simon Fraser University)\n\n
END:VEVENT
END:VCALENDAR