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:20240214T070245Z
LOCATION:Meeting Room C4.11\, Level 4 (Convention Centre)
DTSTART;TZID=Australia/Melbourne:20231213T161500
DTEND;TZID=Australia/Melbourne:20231213T162500
UID:siggraphasia_SIGGRAPH Asia 2023_sess125_papers_769@linklings.com
SUMMARY:Shadow Harmonization for Realistic Compositing
DESCRIPTION:Technical Papers, TOG\n\nLucas Valença and Jinsong Zhang (Univ
 ersité Laval), Michaël Gharbi and Yannick Hold-Geoffroy (Adobe), and Jean-
 François Lalonde (Université Laval)\n\nCompositing virtual objects into re
 al background images requires one to carefully match the scene's camera pa
 rameters, surface geometry, textures, and lighting to obtain plausible ren
 derings.\nRecent learning approaches have shown many scene properties can 
 be estimated from images, resulting in robust automatic single-image compo
 siting systems, but many challenges remain.\nIn particular, interactions b
 etween real and synthetic shadows are not handled gracefully by existing m
 ethods, which typically assume a shadow-free background. \nAs a result, th
 ey tend to generate double shadows when the synthetic object's cast shadow
  overlaps a background shadow, and ignore shadows from the background that
  should be cast onto the synthetic object. \nIn this paper, we present a c
 ompositing method for outdoor scenes that addresses these issues and produ
 ces realistic cast shadows.\nThis requires identifying existing shadows, i
 ncluding soft shadow boundaries, then reasoning about the ambiguity of unk
 nown ground albedo and scene lighting to match the color and intensity of 
 shaded areas.\nUsing supervision from shadow removal and detection dataset
 s, we propose a generative adversarial pipeline and improved composition e
 quations that simultaneously handle both shadow interaction scenarios. \nW
 e evaluate our method on challenging, real outdoor images from multiple di
 stributions and datasets.\nQuantitative and qualitative comparisons show o
 ur approach produces more realistic results than existing alternatives.\n\
 nRegistration Category: Full Access\n\nSession Chair: Michael Gharbi (Adob
 e, MIT)
URL:https://asia.siggraph.org/2023/full-program?id=papers_769&sess=sess125
END:VEVENT
END:VCALENDAR