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_146@linklings.com
SUMMARY:Rhythmic Gesticulator: Rhythm-Aware Co-Speech Gesture Synthesis wi
 th Hierarchical Neural Embeddings
DESCRIPTION:Technical Papers\n\nRhythmic Gesticulator: Rhythm-Aware Co-Spe
 ech Gesture Synthesis with Hierarchical Neural Embeddings\n\nAo, Gao, Lou,
  Chen, Liu\n\nAutomatic synthesis of realistic co-speech gestures is an in
 creasingly important yet challenging task in artificial embodied agent cre
 ation. Previous systems mainly focus on generating gestures in an end-to-e
 nd manner, which leads to difficulties in mining the clear rhythm and sema
 ntics due to the complex yet subtle harmony between speech and gestures. W
 e present a novel co-speech gesture synthesis method that achieves convinc
 ing results both on the rhythm and semantics. For the rhythm, our system c
 ontains a robust rhythm-based segmentation pipeline to ensure the temporal
  coherence between the vocalization and gestures explicitly. For the gestu
 re semantics, we devise a mechanism to effectively disentangle both low- a
 nd high-level neural embeddings of speech and motion based on linguistic t
 heory. The high-level embedding corresponds to semantics, while the low-le
 vel embedding relates to subtle variations. Lastly, we build correspondenc
 e between the hierarchical embeddings of the speech and the motion, result
 ing in rhythm- and semantics-aware gesture synthesis. Evaluations with exi
 sting objective metrics, a newly proposed rhythmic metric, and human feedb
 ack show that our method outperforms state-of-the-art systems by a clear m
 argin.\n\nRegistration Category: FULL ACCESS, EXPERIENCE PLUS ACCESS, EXPE
 RIENCE ACCESS, TRADE EXHIBITOR\n\nLanguage: ENGLISH\n\nFormat: IN-PERSON
URL:https://sa2022.siggraph.org/en/full-program/?id=papers_146&sess=sess15
 3
END:VEVENT
END:VCALENDAR