BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Asia/Seoul
X-LIC-LOCATION:Asia/Seoul
BEGIN:STANDARD
TZOFFSETFROM:+0900
TZOFFSETTO:+0900
TZNAME:KST
DTSTART:18871231T000000
DTSTART:19881009T020000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20230103T035307Z
LOCATION:Auditorium\, Level 5\, West Wing
DTSTART;TZID=Asia/Seoul:20221206T100000
DTEND;TZID=Asia/Seoul:20221206T120000
UID:siggraphasia_SIGGRAPH Asia 2022_sess153_papers_370@linklings.com
SUMMARY:Neural Cloth Simulation
DESCRIPTION:Technical Papers\n\nNeural Cloth Simulation\n\nBertiche, Madad
 i, Escalera\n\nWe present a general framework for the garment animation pr
 oblem through unsupervised deep learning inspired in physically based simu
 lation. Existing trends in the literature already explore this possibility
 . Nonetheless, these approaches do not handle cloth dynamics. Here, we pro
 pose the first methodology able to learn realistic cloth dynamics unsuperv
 isedly, and henceforth, a general formulation for neural cloth simulation.
  The key to achieve this is to adapt an existing optimization scheme for m
 otion from simulation based methodologies to deep learning. Then, analyzin
 g the nature of the problem, we devise an architecture able to automatical
 ly disentangle static and dynamic cloth subspaces by design. We will show 
 how this improves model performance. Additionally, this opens the possibil
 ity of a novel motion augmentation technique that greatly improves general
 ization. Finally, we show it also allows to control the level of motion in
  the predictions. This is a useful, never seen before, tool for artists. W
 e provide of detailed analysis of the problem to establish the bases of ne
 ural cloth simulation and guide future research into the specifics of this
  domain.\n\nRegistration Category: FULL ACCESS, EXPERIENCE PLUS ACCESS, EX
 PERIENCE ACCESS, TRADE EXHIBITOR\n\nLanguage: ENGLISH\n\nFormat: IN-PERSON
URL:https://sa2022.siggraph.org/en/full-program/?id=papers_370&sess=sess15
 3
END:VEVENT
END:VCALENDAR