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:20260417T190159Z
LOCATION:West Building\, Rooms 220-222
DTSTART;TZID=America/Los_Angeles:20250813T140000
DTEND;TZID=America/Los_Angeles:20250813T153000
UID:siggraph_SIGGRAPH 2025_sess154@linklings.com
SUMMARY:Rigging & Interaction
DESCRIPTION:SkillMimic-V2: Learning Robust and Generalizable Interaction S
 kills from Sparse and Noisy Demonstrations\n\nThis work addresses the chal
 lenge of learning robust interaction skills from limited demonstrations. B
 y introducing novel data augmentation techniques for skill transitions and
  recovery patterns, combined with enhanced reinforcement imitation learnin
 g methods, we achieve superior performance in lear...\n\n\nRunyi Yu (Hong 
 Kong University of Science and Technology, Shanghai Aritificial Intelligen
 ce Laboratory); Yinhuai Wang, Qihan Zhao, and Hok Wai Tsui (Hong Kong Univ
 ersity of Science and Technology); Jingbo Wang (Shanghai Aritificial Intel
 ligence Laboratory); and Ping Tan and Qifeng Chen (Hong Kong University of
  Science and Technology)\n---------------------\nRigging & Interaction - I
 nteractive Discussion\n\nAfter the summary presentations, attendees will p
 articipate 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 instig
 ate further conversation such...\n\n---------------------\nMulti-Person In
 teraction Generation from Two-Person Motion Priors\n\nGenerate exciting mu
 lti-character interactions, such as team fights, with our training-free me
 thod! Multi-character interactions can be decomposed into multiple two-per
 son interactions using a directed graph, which enables repurposing large p
 re-trained two-character motion synthesis models without a...\n\n\nWenning
  Xu, Shiyu Fan, Paul Henderson, and Edmond S. L. Ho (University of Glasgow
 )\n---------------------\nRigAnything: Template-Free Autoregressive Riggin
 g for Diverse 3D Assets\n\nRigAnything is a transformer-based model that a
 utoregressively generates 3D rigging without templates. It sequentially pr
 edicts joints and skeleton topology while assigning skinning weights, work
 ing on objects in any pose. It’s 20× faster than existing methods, complet
 ing rigging in under 2 se...\n\n\nIsabella Liu (University of California S
 an Diego); Zhan Xu, Yifan Wang, and Hao Tan (Adobe Research); Zexiang Xu (
 Hillbot Inc.); Xiaolong Wang (University of California San Diego); Hao Su 
 (University of California San Diego, Hillbot Inc.); and Zifan Shi (Adobe R
 esearch)\n---------------------\nLarge-Scale Multi-Character Interaction S
 ynthesis\n\nThis work introduces a conditional generative framework for la
 rge-scale multi-character interaction synthesis by facilitating natural in
 teractive motions and transitions where characters are coordinated for new
  interactive partners, proposing a coordinatable multi-character interacti
 on space for int...\n\n\nZiyi Chang (Durham University); He Wang (UCL Cent
 re for Artificial Intelligence, Department of Computer Science, University
  College London (UCL)); and George Koulieris and Hubert Shum (Durham Unive
 rsity)\n---------------------\nAnymate: A Dataset and Baselines for Learni
 ng 3D Object Rigging\n\nWe present the Anymate Dataset, a large-scale data
 set of 230K 3D assets paired with expert-crafted rigging and skinning info
 rmation---70 times larger than existing datasets. Using this dataset, we p
 ropose a learning-based auto-rigging framework with three sequential modul
 es for joint, connectivity, ...\n\n\nYufan Deng, Yuhao Zhang, and Chen Gen
 g (Stanford University); Shangzhe Wu (Stanford University, University of C
 ambridge); and Jiajun Wu (Stanford University)\n---------------------\nOne
  Model to Rig Them All: Diverse Skeleton Rigging with UniRig\n\nManual 3D 
 rigging is slow. UniRig introduces a unified learning framework for automa
 tic skeletal rigging. Trained on our large, diverse Rig-XL dataset, it use
 s an autoregressive model and cross-attention to accurately rig various ch
 aracters and objects, significantly outperforming prior methods and ...\n\
 n\nJia-Peng Zhang, Cheng-Feng Pu, and Meng-Hao Guo (CS Dept, Tsinghua Univ
 ersity); Yan-Pei Cao (VAST); and Shi-Min Hu (CS Dept, Tsinghua University)
 \n\nInterest Area: Research & Education\n\nRecording: Livestreamed, Not Li
 vestreamed, Recorded, Not Recorded\n\nRegistration Category: Full Conferen
 ce, Virtual Access, Wednesday
END:VEVENT
END:VCALENDAR