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:20240214T070242Z
LOCATION:Darling Harbour Theatre\, Level 2 (Convention Centre)
DTSTART;TZID=Australia/Melbourne:20231212T093000
DTEND;TZID=Australia/Melbourne:20231212T124500
UID:siggraphasia_SIGGRAPH Asia 2023_sess209_papers_769@linklings.com
SUMMARY:Shadow Harmonization for Realistic Compositing
DESCRIPTION:Technical Papers\n\nLucas Valença and Jinsong Zhang (Universit
 é Laval), Michaël Gharbi and Yannick Hold-Geoffroy (Adobe), and Jean-Franç
 ois Lalonde (Université Laval)\n\nCompositing virtual objects into real ba
 ckground images requires one to carefully match the scene's camera paramet
 ers, surface geometry, textures, and lighting to obtain plausible renderin
 gs.\nRecent learning approaches have shown many scene properties can be es
 timated from images, resulting in robust automatic single-image compositin
 g systems, but many challenges remain.\nIn particular, interactions betwee
 n real and synthetic shadows are not handled gracefully by existing method
 s, which typically assume a shadow-free background. \nAs a result, they te
 nd to generate double shadows when the synthetic object's cast shadow over
 laps a background shadow, and ignore shadows from the background that shou
 ld be cast onto the synthetic object. \nIn this paper, we present a compos
 iting method for outdoor scenes that addresses these issues and produces r
 ealistic cast shadows.\nThis requires identifying existing shadows, includ
 ing soft shadow boundaries, then reasoning about the ambiguity of unknown 
 ground albedo and scene lighting to match the color and intensity of shade
 d areas.\nUsing supervision from shadow removal and detection datasets, we
  propose a generative adversarial pipeline and improved composition equati
 ons that simultaneously handle both shadow interaction scenarios. \nWe eva
 luate our method on challenging, real outdoor images from multiple distrib
 utions and datasets.\nQuantitative and qualitative comparisons show our ap
 proach produces more realistic results than existing alternatives.\n\nRegi
 stration Category: Full Access, Enhanced Access, Trade Exhibitor, Experien
 ce Hall Exhibitor
URL:https://asia.siggraph.org/2023/full-program?id=papers_769&sess=sess209
END:VEVENT
END:VCALENDAR