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:20260114T163729Z
LOCATION:Meeting Room C4.8\, Level 4 (Convention Centre)
DTSTART;TZID=Australia/Melbourne:20231212T160000
DTEND;TZID=Australia/Melbourne:20231212T161000
UID:siggraphasia_SIGGRAPH Asia 2023_sess160_papers_998@linklings.com
SUMMARY:MOCHA: Real-Time Motion Characterization via Context Matching
DESCRIPTION:Deok-Kyeong Jang (KAIST, MOVIN Inc.); Yuting Ye (Meta); Jungda
 m Won (Seoul National University); and Sung-Hee Lee (KAIST)\n\nTransformin
 g neutral, characterless input motions to embody the distinct style of a n
 otable character in real time is highly compelling for character animation
 . This paper introduces MOCHA, a novel online motion characterization fram
 ework that transfers both motion styles and body proportions from a target
  character to an input source motion. MOCHA begins by encoding the input m
 otion into a motion feature that structures the body part topology and cap
 tures motion dependencies for effective characterization. Central to our f
 ramework is the Neural Context Matcher, which generates a motion feature f
 or the target character with the most similar context to the input motion 
 feature. The conditioned autoregressive model of the Neural Context Matche
 r can produce temporally coherent character features in each time frame. T
 o generate the final characterized pose, our Characterizer network incorpo
 rates the characteristic aspects of the target motion feature into the inp
 ut motion feature while preserving its context. This is achieved through a
  transformer model that introduces the adaptive instance normalization and
  context mapping-based cross-attention, effectively injecting the characte
 r feature into the source feature. We validate the performance of our fram
 ework through comparisons with prior work and an ablation study. Our frame
 work can easily accommodate various applications, including characterizati
 on with only sparse input and real-time characterization. Additionally, we
  contribute a high-quality motion dataset comprising six different charact
 ers performing a range of motions, which can serve as a valuable resource 
 for future research.\n\nRegistration Category: Full Access\n\nSession Chai
 r: Ioannis Karamouzas (University of California Riverside)\n\n
URL:https://asia.siggraph.org/2023/full-program?id=papers_998&sess=sess160
END:VEVENT
END:VCALENDAR