BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Australia/Melbourne
X-LIC-LOCATION:Australia/Melbourne
BEGIN:DAYLIGHT
TZOFFSETFROM:+1000
TZOFFSETTO:+1100
TZNAME:AEDT
DTSTART:19721003T020000
RRULE:FREQ=YEARLY;BYMONTH=4;BYDAY=1SU
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:19721003T020000
TZOFFSETFROM:+1100
TZOFFSETTO:+1000
TZNAME:AEST
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20240214T070245Z
LOCATION:Meeting Room C4.9+C4.10\, Level 4 (Convention Centre)
DTSTART;TZID=Australia/Melbourne:20231213T181300
DTEND;TZID=Australia/Melbourne:20231213T182300
UID:siggraphasia_SIGGRAPH Asia 2023_sess167_papers_443@linklings.com
SUMMARY:Discovering Fatigued Movements for Virtual Character Animation
DESCRIPTION:Technical Communications, Technical Papers\n\nNoshaba Cheema (
 German Research Center for Artificial Intelligence, Max-Planck Institute f
 or Informatics); Rui Xu (German Research Center for Artificial Intelligenc
 e, Saarland University); Nam Hee Kim and Perttu Hämäläinen (Aalto Universi
 ty); Vladislav Golyanik and Marc Habermann (Max-Planck-Institut für Inform
 atik); Christian Theobalt (Max-Planck-Institut für Informatik, Saarland Un
 iversity); and Philipp Slusallek (Saarland University, German Research Cen
 ter for Artificial Intelligence)\n\nVirtual character animation and moveme
 nt synthesis have advanced rapidly during recent years, especially through
  a combination of extensive motion capture datasets and machine learning. 
 A remaining challenge is interactively simulating characters that fatigue 
 when performing extended motions, which is indispensable for the realism o
 f generated animations. However, capturing such movements is problematic, 
 as performing movements like backflips with fatigued variations up to exha
 ustion raises capture cost and risk of injury. Surprisingly, little resear
 ch has been done on faithful fatigue modeling. To address this, we propose
  a deep reinforcement learning-based approach, which---for the first time 
 in literature---generates control policies for full-body physically simula
 ted agents aware of cumulative fatigue. For this, we first leverage Genera
 tive Adversarial Imitation Learning (GAIL) to learn an expert policy for t
 he skill; Second, we learn a fatigue policy by limiting the generated cons
 tant torque bounds based on endurance time to non-linear, state- and time-
 dependent limits in the joint-actuation space using a Three-Compartment Co
 ntroller (3CC) model. Our results demonstrate that agents can adapt to dif
 ferent fatigue and rest rates interactively, and discover realistic recove
 ry strategies without the need for any captured data of fatigued movement.
 \n\nRegistration Category: Full Access\n\nSession Chair: Yoonsang Lee (Han
 yang University)
URL:https://asia.siggraph.org/2023/full-program?id=papers_443&sess=sess167
END:VEVENT
END:VCALENDAR