BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Asia/Tokyo
X-LIC-LOCATION:Asia/Tokyo
BEGIN:STANDARD
TZOFFSETFROM:+0900
TZOFFSETTO:+0900
TZNAME:JST
DTSTART:18871231T000000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260817T171536Z
LOCATION:Hall B5 (2)\, B Block\, Level 5
DTSTART;TZID=Asia/Tokyo:20241206T090000
DTEND;TZID=Asia/Tokyo:20241206T101000
UID:siggraphasia_SIGGRAPH Asia 2024_sess140@linklings.com
SUMMARY:Differentiable Rendering
DESCRIPTION:Each Paper gives a 10 minute presentation.\n\nDifferentiating 
 Variance for Variance-Aware Inverse Rendering\n\nMonte Carlo methods have 
 been widely adopted in physics-based rendering.\n    A key property of a M
 onte Carlo estimator is its variance, which dictates the convergence rate 
 of the estimator.\n    In this paper, we devise a mathematical formulation
  for derivatives of rendering variance with respect to ...\n\n\nKai Yan (U
 niversity of California Irvine, Wētā FX); Vincent Pegoraro, Marc Droske, a
 nd Jiří Vorba (Wētā FX); and Shuang Zhao (University of California Irvine)
 \n---------------------\nDifferentiable Owen Scrambling\n\nQuasi-Monte Car
 lo integration is at the core of rendering. This technique estimates the v
 alue of an integral by evaluating the integrand at well-chosen sample loca
 tions. These sample points are designed to cover the domain as uniformly a
 s possible to achieve better convergence rates than purely rand...\n\n\nBa
 stien Doignies (Université Claude Bernard Lyon, LIRIS); David Coeurjolly, 
 Nicolas Bonneel, and Julie Digne (CNRS, LIRIS); and Jean-Claude Iehl and V
 ictor Ostromoukhov (Université Claude Bernard Lyon, LIRIS)\n--------------
 -------\nNeural Differential Appearance Equations\n\nWe propose a method t
 o reproduce dynamic appearance textures with space-stationary but time-var
 ying visual statistics.\nWhile most previous work decomposes dynamic textu
 res into static appearance and motion, we focus on dynamic appearance that
  results not from motion but variations of fundamental pro...\n\n\nChen Li
 u and Tobias Ritschel (University College London (UCL))\n-----------------
 ----\nDifferentiable Photon Mapping using Generalized Path Gradients\n\nPh
 oton mapping is a fundamental and practical Monte Carlo rendering techniqu
 e for efficiently simulating global illumination effects, especially for c
 austics and specular-diffuse-specular (SDS) paths. In this paper, we prese
 nt the first differentiable rendering method for photon mapping. The core 
 of...\n\n\nJiankai Xing and Zengyu Li (Tsinghua University), Fujun Luan (A
 dobe Research), and Kun Xu (Tsinghua University)\n---------------------\nM
 arkov-Chain Monte Carlo Sampling of Visibility Boundaries for Differentiab
 le Rendering\n\nPhysics-based differentiable rendering requires estimating
  boundary path integrals emerging from the shift of discontinuities (e.g.,
  visibility boundaries). Previously, although the mathematical formulation
  of boundary path integrals has been established, efficient and robust est
 imation of these int...\n\n\nPeiyu Xu (University of California Irvine); S
 ai Bangaru (MIT CSAIL, NVIDIA Research); Tzu-Mao Li (University of Califor
 nia San Diego); and Shuang Zhao (University of California Irvine)\n-------
 --------------\nA Simple Approach to Differentiable Rendering of SDFs\n\nW
 e present a simple algorithm for differentiable rendering of surfaces repr
 esented by Signed Distance Fields (SDF), which makes it easy to integrate 
 rendering into gradient-based optimization pipelines. To tackle visibility
 -related derivatives that make rendering non-differentiable, existing phys
 ica...\n\n\nZichen Wang and Xi Deng (Cornell University), Ziyi Zhang and W
 enzel Jakob (EPFL), and Steve Marschner (Cornell University)\n\nRegistrati
 on Category: Full Access, Full Access Supporter\n\nLanguage Format: Englis
 h Language\n\nSession Chair: Seungyong Lee (POSTECH)
END:VEVENT
END:VCALENDAR