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:20240214T070241Z
LOCATION:Darling Harbour Theatre\, Level 2 (Convention Centre)
DTSTART;TZID=Australia/Melbourne:20231212T093000
DTEND;TZID=Australia/Melbourne:20231212T124500
UID:siggraphasia_SIGGRAPH Asia 2023_sess209_papers_612@linklings.com
SUMMARY:Neural Point-based Volumetric Avatar: Surface-guided Neural Points
  for Efficient and Photorealistic Volumetric Head Avatar
DESCRIPTION:Technical Papers\n\nCong Wang (Tsinghua University); Di Kang, 
 Yan-Pei Cao, Linchao Bao, and Ying Shan (Tencent); and Song-Hai Zhang (Tsi
 nghua University)\n\nRendering photo-realistic and vividly moving human he
 ads is very important for pleasant and immersive experience in AR/VR and v
 ideo conferencing. However, existing methods usually struggle to model cha
 llenging facial regions (e.g., mouth interior, eyes, hair/beard), resultin
 g in unrealistic and blurry results. In this paper, we propose Neural Poin
 t-based Volumetric Avatar (NPVA), which discards predefined connectivity a
 nd hard correspondence imposed by mesh-based methods (i.e. neural points) 
 and adopts neural volume rendering. Specifically, the neural points are co
 nstrained around the surface of the target expression via a high-resolutio
 n UV displacement map, achieving increased modeling capacity and more accu
 rate control. We propose three technical innovations to improve the render
 ing and training efficiency, including a patch-wise depth-guided (shading 
 point) sampling strategy, a lightweight radiance decoding process, and a G
 rid-Error-Patch (GEP) ray sampling strategy during training. By design, ou
 r NPVA can better handle topologically changing regions and thin structure
 s, and can be animated with accurate expression control.\n\nRegistration C
 ategory: Full Access, Enhanced Access, Trade Exhibitor, Experience Hall Ex
 hibitor
URL:https://asia.siggraph.org/2023/full-program?id=papers_612&sess=sess209
END:VEVENT
END:VCALENDAR