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:20240214T070311Z
LOCATION:Meeting Room C4.9+C4.10\, Level 4 (Convention Centre)
DTSTART;TZID=Australia/Melbourne:20231213T152000
DTEND;TZID=Australia/Melbourne:20231213T162500
UID:siggraphasia_SIGGRAPH Asia 2023_sess168@linklings.com
SUMMARY:Applications & Innovations
DESCRIPTION:Technical Papers\n\nProjective Sampling for Differentiable Ren
 dering of Geometry\n\nDiscontinuous visibility changes at object boundarie
 s remain a persistent source of difficulty in the area of differentiable r
 endering. Left untreated, they bias computed gradients so severely that ev
 en basic optimization tasks fail.\n\nPrior path-space methods addressed th
 is bias by decoupling bounda...\n\n\nZiyi Zhang, Nicolas Roussel, and Wenz
 el Jakob (Ecole Polytechnique Fédérale de Lausanne)\n---------------------
 \nExtended Path Space Manifolds for Physically Based Differentiable Render
 ing\n\nPhysically based differentiable rendering has become an increasingl
 y important topic in recent years. A common pipeline computes local color 
 derivatives of light paths or pixels with respect to arbitrary scene param
 eters, and enables optimizing or recovering the scene parameters through i
 terative gr...\n\n\nJiankai Xing and Xuejun Hu (Tsinghua University), Fuju
 n Luan (Adobe Research), Ling-Qi Yan (University of California Santa Barba
 ra), and Kun Xu (Tsinghua University)\n---------------------\nWarped-Area 
 Reparameterization of Differential Path Integrals\n\nPhysics-based differe
 ntiable rendering is becoming increasingly crucial for tasks in inverse re
 ndering and machine learning pipelines. To address discontinuities caused 
 by geometric boundaries and occlusion, two classes of methods have been pr
 oposed: 1) the edge sampling methods that directly sample...\n\n\nPeiyu Xu
  (University of California Irvine), Sai Bangaru (MIT CSAIL), Tzu-Mao Li (U
 niversity of California San Diego), and Shuang Zhao (University of Califor
 nia Irvine)\n---------------------\nJoint Sampling and Optimisation for In
 verse Rendering\n\nWhen dealing with difficult inverse problems such as in
 verse rendering, using Monte Carlo estimated gradients to optimise paramet
 ers can slow down convergence due to variance. Averaging many gradient sam
 ples in each iteration reduces this variance trivially. However, for probl
 ems that require thousa...\n\n\nMartin Balint, Karol Myszkowski, Hans-Pete
 r Seidel, and Gurprit Singh (Max Planck Institute for Informatics)\n------
 ---------------\nAmortizing Samples in Physics-Based Inverse Rendering usi
 ng ReSTIR\n\nRecently, great progress has been made in physics-based diffe
 rentiable rendering. Existing differentiable rendering techniques typicall
 y focus on static scenes, but during inverse rendering—a key application f
 or differentiable rendering—the scene is updated dynamically by each gradi
 ent s...\n\n\nYu-Chen Wang (University of California Irvine), Chris Wyman 
 and Lifan Wu (NVIDIA), and Shuang Zhao (University of California Irvine)\n
 \nRegistration Category: Full Access\n\nSession Chair: Soo-Mi Choi (Sejong
  University)
END:VEVENT
END:VCALENDAR