BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Australia/Melbourne
X-LIC-LOCATION:Australia/Melbourne
BEGIN:DAYLIGHT
TZOFFSETFROM:+1000
TZOFFSETTO:+1100
TZNAME:AEDT
DTSTART:19721003T020000
RRULE:FREQ=YEARLY;BYMONTH=4;BYDAY=1SU
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:19721003T020000
TZOFFSETFROM:+1100
TZOFFSETTO:+1000
TZNAME:AEST
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260114T163654Z
LOCATION:Meeting Room C4.8\, Level 4 (Convention Centre)
DTSTART;TZID=Australia/Melbourne:20231215T104500
DTEND;TZID=Australia/Melbourne:20231215T105500
UID:siggraphasia_SIGGRAPH Asia 2023_sess154_papers_1043@linklings.com
SUMMARY:ActRay: Online Active Ray Sampling for Radiance Fields
DESCRIPTION:Jiangkai Wu, Liming Liu, Yunpeng Tan, Quanlu Jia, Haodan Zhang
 , and Xinggong Zhang (Peking University)\n\nThanks to the high-quality rec
 onstruction and photorealistic rendering, the Neural Radiance Field (NeRF)
  has garnered extensive attention and has been continuously improved. Desp
 ite its high visual quality, the prohibitive training time limits its prac
 tical application. Although significant acceleration has been achieved, it
  is still far from real-time training, due to the need for tens of thousan
 ds of iterations. In this paper, a feasible solution is to reduce the numb
 er of required iterations by always training the rays with the highest los
 s values, instead of the traditional method of training each ray with a un
 iform probability. To this end, we propose an online active ray sampling s
 trategy, ActRay. Specifically, to avoid the substantial overhead of calcul
 ating the actual loss values for all rays in each iteration, a rendering-g
 radient-based loss propagation algorithm is presented to efficiently estim
 ate the loss values. To further narrow the gap between the estimated loss 
 and the actual loss, an online learning algorithm based on the Upper Confi
 dence Bound (UCB) is proposed to control the sampling probability of the r
 ays, thereby compensating for the bias in loss estimation. We evaluate Act
 Ray on both real-world and synthetic scenes, and the promising results sho
 w that it accelerates radiance field training to 6.5x. Besides, we test Ac
 tRay under all kinds of representations of radiance fields (implicit, expl
 icit, and hybrid), demonstrating that it is general and effective to diffe
 rent representations. We believe this work will contribute to the practica
 l application of radiance fields, because it has taken a step closer to re
 al-time radiance field training.\n\nRegistration Category: Full Access\n\n
 Session Chair: Yuchi Huo (Zhejiang University, Korea Advanced Institute of
  Science and Technology)\n\n
URL:https://asia.siggraph.org/2023/full-program?id=papers_1043&sess=sess15
 4
END:VEVENT
END:VCALENDAR