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:20260114T163655Z
LOCATION:Meeting Room C4.9+C4.10\, Level 4 (Convention Centre)
DTSTART;TZID=Australia/Melbourne:20231213T154500
DTEND;TZID=Australia/Melbourne:20231213T155500
UID:siggraphasia_SIGGRAPH Asia 2023_sess168_papers_546@linklings.com
SUMMARY:Extended Path Space Manifolds for Physically Based Differentiable 
 Rendering
DESCRIPTION:Jiankai Xing and Xuejun Hu (Tsinghua University), Fujun Luan (
 Adobe Research), Ling-Qi Yan (University of California Santa Barbara), and
  Kun Xu (Tsinghua University)\n\nPhysically based differentiable rendering
  has become an increasingly important topic in recent years. A common pipe
 line computes local color derivatives of light paths or pixels with respec
 t to arbitrary scene parameters, and enables optimizing or recovering the 
 scene parameters through iterative gradient descent by minimizing the diff
 erence between rendered and target images. However, existing approaches ca
 nnot robustly handle complex illumination effects including reflections, r
 efractions, caustics, shadows, and highlights, especially when the initial
  and target locations of such illumination effects are not close to each o
 ther in the image space.\nTo address this problem, we propose a novel data
  structure named extended path space manifolds. The manifolds are defined 
 in the combined space of path vertices and scene parameters. By enforcing 
 geometric constraints, the path vertices could be implicitly and uniquely 
 determined by perturbed scene parameters. This enables the manifold to tra
 ck specific illumination effects and the corresponding paths, i.e., specul
 ar paths will still be specular paths after scene parameters are perturbed
 . Besides, the path derivatives with respect to scene parameters could be 
 computed by solving small linear systems. \nWe further propose a physicall
 y based differentiable rendering method built upon the theoretical results
  of extended path space manifolds. By incorporating the path derivatives c
 omputed from the manifolds and an optimal transport based loss function, o
 ur method is demonstrated to be more effective and robust than state-of-th
 e-art approaches in inverse rendering applications involving complex illum
 ination effects.\n\nRegistration Category: Full Access\n\nSession Chair: S
 oo-Mi Choi (Sejong University)\n\n
URL:https://asia.siggraph.org/2023/full-program?id=papers_546&sess=sess168
END:VEVENT
END:VCALENDAR
