BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Asia/Hong_Kong
X-LIC-LOCATION:Asia/Hong_Kong
BEGIN:STANDARD
TZOFFSETFROM:+0800
TZOFFSETTO:+0800
TZNAME:HKT
DTSTART:19911015T033000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20251218T030656Z
LOCATION:Meeting Room S426+S427\, Level 4
DTSTART;TZID=Asia/Hong_Kong:20251217T113400
DTEND;TZID=Asia/Hong_Kong:20251217T114500
UID:siggraphasia_SIGGRAPH Asia 2025_sess136_papers_2493@linklings.com
SUMMARY:JoruriPuppet: Learning Tempo-Changing Mechanisms Beyond the Beat f
 or Music-to-Motion Generation with Expressive Metrics
DESCRIPTION:Ran Dong (Chukyo University), Shaowen Ni (Mie University), and
  Xi Yang (Jilin University)\n\nIn music-to-motion generation, the interpla
 y between movements and music tempo variations significantly influences th
 e emotional expressiveness and realism of performances. However, tempo-cha
 nging mechanisms remain underexplored in neural network-based music-to-mot
 ion tasks due to the scarcity of relevant datasets. Therefore, in this pap
 er, we propose to use novel music features explicitly representing tempo v
 ariations, and introduce a dataset, JoruriPuppet, incorporating the Japane
 se traditional Jo-Ha-Kyu principle characterized by expressive tempo chang
 es. Furthermore, we design three metrics to quantitatively evaluate the sy
 nchronization and expressiveness of generated motions. Experiments on our 
 dataset highlight the limitations of SOTA methods in capturing fine-graine
 d tempo changes. We demonstrate that integrating tempo-changing features i
 nto them improves neural network-based music-to-motion performance across 
 existing datasets, validating the general effectiveness and applicability 
 of our research.\n\nRegistration Category: Full Access, Full Access Suppor
 ter\n\nSession Chair: Yinghao Xu (Stanford University)\n\n
END:VEVENT
END:VCALENDAR