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:20230103T035308Z
LOCATION:Room 325-AB\, Level 3\, West Wing
DTSTART;TZID=Asia/Seoul:20221207T110000
DTEND;TZID=Asia/Seoul:20221207T123000
UID:siggraphasia_SIGGRAPH Asia 2022_sess161_papers_219@linklings.com
SUMMARY:Fast Dynamic Radiance Fields with Time-Aware Neural Voxels
DESCRIPTION:Technical Communications, Technical Papers\n\nFast Dynamic Rad
 iance Fields with Time-Aware Neural Voxels\n\nFang, Yi, Wang, Xie, Zhang..
 .\n\nNeural radiance fields (NeRF) have shown great success in modeling 3D
  scenes and synthesizing novel-view images. However, most previous NeRF me
 thods take much time to optimize one single scene. Explicit data structure
 s, e.g. voxel features, show great potential to accelerate the training pr
 ocess. However, voxel features face two big challenges to be applied to dy
 namic scenes, i.e. modeling temporal information and capturing different s
 cales of point motions. We propose a radiance field framework by represent
 ing scenes with time-aware voxel features, named as TiNeuVox. A tiny coord
 inate deformation network is introduced to model coarse motion trajectorie
 s and temporal information is further enhanced in the radiance network. A 
 multi-distance interpolation method is proposed and applied on voxel featu
 res to model both small and large motions. Our framework significantly acc
 elerates the optimization of dynamic radiance fields while maintaining hig
 h rendering quality. Empirical evaluation is performed on both synthetic a
 nd real scenes. Our TiNeuVox completes training with only 8 minutes and 8-
 MB storage cost while showing similar or even better rendering performance
  than previous dynamic NeRF methods.\n\nRegistration Category: FULL ACCESS
 , ON-DEMAND ACCESS\n\nLanguage: ENGLISH\n\nFormat: IN-PERSON, ON-DEMAND
URL:https://sa2022.siggraph.org/en/full-program/?id=papers_219&sess=sess16
 1
END:VEVENT
END:VCALENDAR