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:20240214T070249Z
LOCATION:Meeting Room C4.8\, Level 4 (Convention Centre)
DTSTART;TZID=Australia/Melbourne:20231215T101500
DTEND;TZID=Australia/Melbourne:20231215T103000
UID:siggraphasia_SIGGRAPH Asia 2023_sess154_papers_774@linklings.com
SUMMARY:Adaptive Shells for Efficient Neural Radiance Field Rendering
DESCRIPTION:Technical Papers\n\nZian Wang and Tianchang Shen (NVIDIA, Univ
 ersity of Toronto); Merlin Nimier-David and Nicholas Sharp (NVIDIA); Jun G
 ao (NVIDIA, University of Toronto); Alexander Keller (NVIDIA); Sanja Fidle
 r (NVIDIA, University of Toronto); and Thomas Müller and Zan Gojcic (NVIDI
 A)\n\nNeural radiance fields achieve unprecedented quality for novel view 
 synthesis, but their volumetric formulation remains expensive, requiring a
  huge number of samples to render high-resolution images. Volumetric encod
 ings are essential to represent fuzzy geometry such as foliage and hair, a
 nd they are well-suited for stochastic optimization. Yet, many scenes ulti
 mately consist largely of solid surfaces which can be accurately rendered 
 by a single sample per pixel. Based on this insight, we propose a neural r
 adiance formulation that smoothly transitions between volumetric- and surf
 ace-based rendering, greatly accelerating rendering speed and even improvi
 ng visual fidelity. Our method constructs an explicit mesh envelope which 
 spatially bounds a neural volumetric representation. In solid regions, the
  envelope nearly converges to a surface and can often be rendered with a s
 ingle sample. To this end, we generalize the NeuS formulation with a learn
 ed spatially-varying kernel size which encodes the spread of the density, 
 fitting a wide kernel to volume-like regions and a tight kernel to surface
 -like regions. We then extract an explicit mesh of a narrow band around th
 e surface, with width determined by the kernel size, and fine-tune the rad
 iance field within this band. At inference time, we cast rays against the 
 mesh and evaluate the radiance field only within the enclosed region, grea
 tly reducing the number of samples required. Experiments show that our app
 roach enables efficient rendering at very high fidelity. We also demonstra
 te that the extracted envelope enables downstream applications such as ani
 mation and simulation.\n\nRegistration Category: Full Access\n\nSession Ch
 air: Yuchi Huo (Zhejiang University, Korea Advanced Institute of Science a
 nd Technology)
URL:https://asia.siggraph.org/2023/full-program?id=papers_774&sess=sess154
END:VEVENT
END:VCALENDAR