BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Asia/Tokyo
X-LIC-LOCATION:Asia/Tokyo
BEGIN:STANDARD
TZOFFSETFROM:+0900
TZOFFSETTO:+0900
TZNAME:JST
DTSTART:18871231T000000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260817T171531Z
LOCATION:Hall B7 (1)\, B Block\, Level 7
DTSTART;TZID=Asia/Tokyo:20241205T114300
DTEND;TZID=Asia/Tokyo:20241205T115400
UID:siggraphasia_SIGGRAPH Asia 2024_sess129_papers_1174@linklings.com
SUMMARY:ELMO: Enhanced Real-time LiDAR Motion Capture through Upsampling
DESCRIPTION:Deok-Kyeong Jang (MOVIN Inc.); Dongseok Yang (MOVIN Inc., KAIS
 T); Deok-Yun Jang (MOVIN Inc., GIST); Byeoli Choi (MOVIN Inc., KAIST); Don
 ghoon Shin (MOVIN Inc.); and Sung-Hee Lee (KAIST)\n\nThis paper introduces
  ELMO, a real-time upsampling motion capture framework designed for a sing
 le LiDAR sensor. Modeled as a conditional autoregressive transformer-based
  upsampling motion generator, ELMO achieves 60 fps motion capture from a 2
 0 fps LiDAR point cloud sequence. The key feature of ELMO is the coupling 
 of the self-attention mechanism with thoughtfully designed embedding modul
 es for motion and point clouds, significantly elevating the motion quality
 . \nTo facilitate accurate motion capture, we develop a one-time skeleton 
 calibration model capable of predicting user skeleton offsets from a singl
 e-frame point cloud. Additionally, we introduce a novel data augmentation 
 technique utilizing a LiDAR simulator, which enhances global root tracking
  to improve environmental understanding.\nTo demonstrate the effectiveness
  of our method, we compare ELMO with state-of-the-art methods in both imag
 e-based and point cloud-based motion capture. We further conduct an ablati
 on study to validate our design principles. \nELMO's fast inference time m
 akes it well-suited for real-time applications, exemplified in our demo vi
 deo featuring live streaming and interactive gaming scenarios. \nFurthermo
 re, we contribute a high-quality LiDAR-mocap synchronized dataset comprisi
 ng 20 different subjects performing a range of motions, which can serve as
  a valuable resource for future research.\n\nRegistration Category: Full A
 ccess, Full Access Supporter\n\nLanguage Format: English Language\n\nSessi
 on Chair: Yuting Ye (Reality Labs Research, Meta; Meta)\n\n
URL:https://asia.siggraph.org/2024/program/?id=papers_1174&sess=sess129
END:VEVENT
END:VCALENDAR