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:20240214T070248Z
LOCATION:Meeting Room C4.8\, Level 4 (Convention Centre)
DTSTART;TZID=Australia/Melbourne:20231214T165800
DTEND;TZID=Australia/Melbourne:20231214T170800
UID:siggraphasia_SIGGRAPH Asia 2023_sess166_papers_486@linklings.com
SUMMARY:Neural Motion Graph
DESCRIPTION:Technical Communications, Technical Papers\n\nHongyu Tao, Shua
 iying Hou, Changqing Zou, Hujun Bao, and Weiwei Xu (Zhejiang University)\n
 \nDeep learning techniques have been employed to design a controllable hum
 an motion synthesizer. Despite their potential, however, designing a neura
 l network-based motion synthesis that enables flexible user interaction, f
 ine-grained controllability, and the support of new types of motions at re
 duced time and space consumption costs remains a challenge. In this paper,
  we propose a novel approach, a neural motion graph, that addresses the ch
 allenge by enabling scalability to new motions while using compact neural 
 networks. Our approach represents each type of motion with a separate neur
 al node to reduce the cost of adding new motion types. In addition, design
 ing a separate neural node for each motion type enables task-specific cont
 rol strategies and has greater potential to achieve a high-quality synthes
 is of complex motions, such as the Mongolian dance. Furthermore, a single 
 transition network, which acts as neural edges, is used to model the trans
 ition between two motion nodes. The transition network is designed with a 
 lightweight control module to achieve a fine-grained response to user cont
 rol signals. Overall, the design choice makes the neural motion graph high
 ly controllable and scalable. In addition to being fully flexible to user 
 interaction through high-level and fine-grained user-control signals, our 
 experimental and subjective evaluation results demonstrate that our propos
 ed approach, neural motion graph, outperforms state-of-the-art human motio
 n synthesis methods in terms of the quality of controlled motion generatio
 n.\n\nRegistration Category: Full Access\n\nSession Chair: Seungbae Bang (
 Amazon, University of Toronto)
URL:https://asia.siggraph.org/2023/full-program?id=papers_486&sess=sess166
END:VEVENT
END:VCALENDAR