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:20250110T023312Z
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:Technical Papers\n\nDeok-Kyeong Jang (MOVIN Inc.); Dongseok Ya
 ng (MOVIN Inc., KAIST); Deok-Yun Jang (MOVIN Inc., GIST); Byeoli Choi (MOV
 IN Inc., KAIST); Donghoon Shin (MOVIN Inc.); and Sung-Hee Lee (KAIST)\n\nT
 his paper introduces ELMO, a real-time upsampling motion capture framework
  designed for a single LiDAR sensor. Modeled as a conditional autoregressi
 ve transformer-based upsampling motion generator, ELMO achieves 60 fps mot
 ion capture from a 20 fps LiDAR point cloud sequence. The key feature of E
 LMO is the coupling of the self-attention mechanism with thoughtfully desi
 gned embedding modules for motion and point clouds, significantly elevatin
 g the motion quality. \nTo facilitate accurate motion capture, we develop 
 a one-time skeleton calibration model capable of predicting user skeleton 
 offsets from a single-frame point cloud. Additionally, we introduce a nove
 l data augmentation technique utilizing a LiDAR simulator, which enhances 
 global root tracking to improve environmental understanding.\nTo demonstra
 te the effectiveness of our method, we compare ELMO with state-of-the-art 
 methods in both image-based and point cloud-based motion capture. We furth
 er conduct an ablation study to validate our design principles. \nELMO's f
 ast inference time makes it well-suited for real-time applications, exempl
 ified in our demo video featuring live streaming and interactive gaming sc
 enarios. \nFurthermore, we contribute a high-quality LiDAR-mocap synchroni
 zed dataset comprising 20 different subjects performing a range of motions
 , which can serve as a valuable resource for future research.\n\nRegistrat
 ion Category: Full Access, Full Access Supporter\n\nLanguage Format: Engli
 sh Language\n\nSession Chair: Yuting Ye (Reality Labs Research, Meta; Meta
 )
URL:https://asia.siggraph.org/2024/program/?id=papers_1174&sess=sess129
END:VEVENT
END:VCALENDAR