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:20260114T163649Z
LOCATION:Meeting Room C4.11\, Level 4 (Convention Centre)
DTSTART;TZID=Australia/Melbourne:20231215T160000
DTEND;TZID=Australia/Melbourne:20231215T161000
UID:siggraphasia_SIGGRAPH Asia 2023_sess139_papers_240@linklings.com
SUMMARY:A Locality-based Neural Solver for Optical Motion Capture
DESCRIPTION:Xiaoyu Pan and Bowen Zheng (State Key Laboratory of CAD & CG, 
 Zhejiang University; ZJU-Tencent Game and Intelligent Graphics Innovation 
 Technology Joint Lab); Xinwei Jiang, Guanglong Xu, Xianli Gu, and Jingxian
 g Li (Tencent Games Digital Content Technology Center); Qilong Kou (Tencen
 t Technology (Shenzhen) Co., LTD); He Wang (University College London (UCL
 )); Tianjia Shao and Kun Zhou (State Key Laboratory of CAD & CG, Zhejiang 
 University); and Xiaogang Jin (State Key Laboratory of CAD & CG, Zhejiang 
 University; ZJU-Tencent Game and Intelligent Graphics Innovation Technolog
 y Joint Lab)\n\nWe present a novel locality-based learning method for clea
 ning and solving optical motion capture data. Given noisy marker data, we 
 propose a new heterogeneous graph neural network which treats markers and 
 joints as different types of nodes, and uses graph convolution operations 
 to extract the local features of markers and joints and transform them to 
 clean motions. To deal with anomaly markers (e.g. missing or with big trac
 king errors), the key insight is that a marker motion show strong correlat
 ions with the motions of its immediate neighboring markers but less so wit
 h other markers, a.k.a. locality, which enables us to fill missing markers
  (e.g. due to occlusion). Additionally, we also identify marker outliers d
 ue to tracking errors by investigating their acceleration profiles. Finall
 y, we propose a training regime based on representation learning and data 
 augmentation, by training the model on data with masking. The masking sche
 mes aim to mimic the missing and noisy markers often observed in the real 
 data. Finally, we show that our method achieves high accuracy on multiple 
 metrics across various datasets. Extensive comparison shows our method out
 performs state-of-the-art methods in terms of prediction accuracy of occlu
 ded marker position error by approximately 20%, which leads to a further e
 rror reduction on the reconstructed joint rotations and positions by 30%. 
 The code and data for this paper are available at github.com/localmocap/Lo
 calMoCap.\n\nRegistration Category: Full Access\n\nSession Chair: Yuting Y
 e (Reality Labs Research, Meta; Meta)\n\n
URL:https://asia.siggraph.org/2023/full-program?id=papers_240&sess=sess139
END:VEVENT
END:VCALENDAR