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:20240214T070250Z
LOCATION:Meeting Room C4.8\, Level 4 (Convention Centre)
DTSTART;TZID=Australia/Melbourne:20231215T111500
DTEND;TZID=Australia/Melbourne:20231215T112500
UID:siggraphasia_SIGGRAPH Asia 2023_sess156_papers_458@linklings.com
SUMMARY:Intrinsic Harmonization for Illumination-Aware Image Compositing
DESCRIPTION:Technical Communications, Technical Papers\n\nChris Careaga, S
 . Mahdi H. Miangoleh, and Yağız Aksoy (Simon Fraser University)\n\nDespite
  significant advancements in network-based image harmonization techniques,
  there still exists a domain gap between training pairs and real-world com
 posites encountered during inference. Most existing methods are trained to
  reverse global edits made on segmented image regions, which fail to accur
 ately capture the lighting inconsistencies between the foreground and back
 ground commonly found in composited images. In this work, we introduce a s
 elf-supervised illumination harmonization approach formulated in the intri
 nsic image domain. First, we estimate a simple global lighting model from 
 mid-level vision representations to generate a rough shading for the foreg
 round region. A network then refines this inferred shading to generate a h
 armonious re-shading that aligns with the background scene. In order to ma
 tch the color appearance of the foreground and background, we utilize idea
 s from prior harmonization approaches to perform global image edits in the
  albedo domain. To validate the effectiveness of our approach, we present 
 results on challenging real-world composites and conduct a user study to o
 bjectively measure the enhanced realism achieved compared to state-of-the-
 art harmonization methods.\n\nRegistration Category: Full Access\n\nSessio
 n Chair: Sergi Pujades (National Institute for Research in Computer Scienc
 e and Automation (INRIA), Université Grenoble Alpes)
URL:https://asia.siggraph.org/2023/full-program?id=papers_458&sess=sess156
END:VEVENT
END:VCALENDAR