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:20260114T163644Z
LOCATION:Meeting Room C4.8\, Level 4 (Convention Centre)
DTSTART;TZID=Australia/Melbourne:20231213T145500
DTEND;TZID=Australia/Melbourne:20231213T150500
UID:siggraphasia_SIGGRAPH Asia 2023_sess144_papers_384@linklings.com
SUMMARY:PSDR-Room: Single Photo to Scene using Differentiable Rendering
DESCRIPTION:Kai Yan (University of California, Irvine; Adobe Research); Fu
 jun Luan, Miloš Hašan, Thibault Groueix, and Valentin Deschaintre (Adobe R
 esearch); and Shuang Zhao (University of California, Irvine)\n\nA 3D digit
 al scene composes many components: lights, materials and geometries, inter
 acting to reach the desired appearance. Staging such a scene is time-consu
 ming and requires both artistic and technical skills. In this work, we pro
 pose a system allowing to optimize lighting as well as the pose and materi
 al of individual objects to match a target image of a room scene, with min
 imal user input.\n    To this end, we leverage a recent path-space differe
 ntiable rendering approach that provides unbiased gradients of the renderi
 ng with respect to geometry, lighting, and procedural materials, allowing 
 us to optimize all of these components using gradient descent to visually 
 match the input photo appearance.\n    We use recent single-image scene un
 derstanding methods to initialize the optimization and search for appropri
 ate 3D models and materials. We evaluate our method on real photographs of
  indoor scenes and demonstrate the editability of the resulting scene comp
 onents.\n\nRegistration Category: Full Access\n\nSession Chair: Oded Stein
  (Technion, University of Southern California)\n\n
URL:https://asia.siggraph.org/2023/full-program?id=papers_384&sess=sess144
END:VEVENT
END:VCALENDAR