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:20251218T030653Z
LOCATION:Meeting Room S421\, Level 4
DTSTART;TZID=Asia/Hong_Kong:20251216T153300
DTEND;TZID=Asia/Hong_Kong:20251216T154400
UID:siggraphasia_SIGGRAPH Asia 2025_sess124_papers_1682@linklings.com
SUMMARY:4DSloMo: 4D Reconstruction for High Speed Scene with Asynchronous 
 Capture
DESCRIPTION:Yutian Chen (Shanghai Artificial Intelligence Laboratory, Chin
 ese University of Hong Kong); Shi Guo (Shanghai Artificial Intelligence La
 boratory); Tianshuo Yang (University of Hong Kong); Lihe Ding and Xiuyuan 
 Yu (Chinese University of Hong Kong); Jinwei Gu (NVIDIA); and Tianfan Xue 
 (Chinese University of Hong Kong)\n\nReconstructing fast-dynamic scenes fr
 om multi-view videos is crucial for high-speed motion analysis and realist
 ic 4D reconstruction. However, the majority of 4D capture systems are limi
 ted to frame rates below 30 FPS (frames per second), and a direct 4D recon
 struction of high-speed motion from low FPS input may lead to undesirable 
 results. In this work, we propose a high-speed 4D capturing system only us
 ing low FPS cameras, through novel capturing and processing modules. On th
 e capturing side, we propose an asynchronous capture scheme that increases
  the effective frame rate by staggering the start times of cameras. By gro
 uping cameras and leveraging a base frame rate of 25 FPS, our method achie
 ves an equivalent frame rate of 100–200 FPS without requiring specialized 
 high-speed cameras. On processing side, we also propose a novel generative
  model to fix artifacts caused by 4D sparse-view reconstruction, as asynch
 rony reduces the number of viewpoints at each timestamp. Specifically, we 
 propose to train a video-diffusion-based artifact-fix model for sparse 4D 
 reconstruction, which refines missing details, maintains temporal consiste
 ncy, and improves overall reconstruction quality. Experimental results dem
 onstrate that our method significantly enhances high-speed 4D reconstructi
 on compared to synchronized capture.\n\nRegistration Category: Full Access
 , Full Access Supporter\n\nSession Chair: Yan-Pei Cao (VAST)\n\n
END:VEVENT
END:VCALENDAR