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:20260114T163641Z
LOCATION:Meeting Room C4.8\, Level 4 (Convention Centre)
DTSTART;TZID=Australia/Melbourne:20231212T171000
DTEND;TZID=Australia/Melbourne:20231212T172000
UID:siggraphasia_SIGGRAPH Asia 2023_sess142_papers_446@linklings.com
SUMMARY:Input-Dependent Uncorrelated Weighting for Monte Carlo Denoising
DESCRIPTION:Jonghee Back (Gwangju Institute of Science and Technology), Bi
 nh-Son Hua (Trinity College Dublin), Toshiya Hachisuka (University of Wate
 rloo), and Bochang Moon (Gwangju Institute of Science and Technology)\n\nI
 mage-space denoising techniques have been widely employed in Monte Carlo r
 endering, typically blending neighboring pixel estimates using a denoising
  kernel. It is widely recognized that a kernel should be adapted to charac
 teristics of the input pixel estimates in order to ensure robustness to di
 verse image features and amount of noise. Denoising with such an input-dep
 endent kernel, however, can introduce a bias that makes the denoised estim
 ate even less accurate than the noisy input estimate. Consequently, it has
  been considered essential to balance the bias introduced by denoising and
  the reduction of noise. We propose a new framework to define an input-dep
 endent kernel that departs from the existing approaches based on error est
 imation or supervised learning. Rather than seeking an optimal bias-noise 
 balance as in those existing approaches, we propose to constrain the amoun
 t of bias introduced by denoising. Such a constraint is made possible by t
 he concept of uncorrelated statistics, which has never been applied for de
 noising. By designing an input-dependent kernel with uncorrelated weights 
 against the input pixel estimates, our denoising kernel can reduce data-de
 pendent noise with a negligible amount of bias in most cases. We demonstra
 te the effectiveness of our method for various scenes.\n\nRegistration Cat
 egory: Full Access\n\nSession Chair: Michael Gharbi (Reve AI, Massachusett
 s Institute of Technology (MIT))\n\n
URL:https://asia.siggraph.org/2023/full-program?id=papers_446&sess=sess142
END:VEVENT
END:VCALENDAR