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:20230103T035306Z
LOCATION:Auditorium\, Level 5\, West Wing
DTSTART;TZID=Asia/Seoul:20221206T100000
DTEND;TZID=Asia/Seoul:20221206T120000
UID:siggraphasia_SIGGRAPH Asia 2022_sess153_papers_112@linklings.com
SUMMARY:Efficient Neural Radiance Fields for Interactive Free-viewpoint Vi
 deo
DESCRIPTION:Technical Papers\n\nEfficient Neural Radiance Fields for Inter
 active Free-viewpoint Video\n\nLin, Peng, Xu, Yan, Shuai...\n\nThis paper 
 aims to tackle the challenge of efficiently producing interactive free-vie
 wpoint videos. \nSome recent works equip neural radiance fields with image
  encoders, enabling them to generalize across scenes. When processing dyna
 mic scenes, they can simply treat each video frame as an individual scene 
 and perform novel view synthesis to generate free-viewpoint videos. Howeve
 r, their rendering process is slow and cannot support interactive applicat
 ions. \nA major factor is that they sample lots of points in empty space w
 hen inferring radiance fields. \nWe propose a novel scene representation, 
 called ENeRF, for the fast creation of interactive free-viewpoint videos. 
 Specifically, given multi-view images at one frame, we first build the cas
 cade cost volume to predict the coarse geometry of the scene. The coarse g
 eometry allows us to sample few points near the scene surface, thereby sig
 nificantly improving the rendering speed. This process is fully differenti
 able, enabling us to jointly learn the depth prediction and radiance field
  networks from RGB images. Experiments show that our approach exhibits com
 petitive performance on the DTU, NeRF Synthetic, Real Forward-facing, ZJU-
 MoCap, and DynamicCap datasets while being at least 60 times faster than p
 revious generalizable radiance field methods. We demonstrate the capabilit
 y of our method to synthesize novel views of human performers in real-time
 .  The code is available at https://zju3dv.github.io/enerf/.\n\nRegistrati
 on Category: FULL ACCESS, EXPERIENCE PLUS ACCESS, EXPERIENCE ACCESS, TRADE
  EXHIBITOR\n\nLanguage: ENGLISH\n\nFormat: IN-PERSON
URL:https://sa2022.siggraph.org/en/full-program/?id=papers_112&sess=sess15
 3
END:VEVENT
END:VCALENDAR