BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Asia/Tokyo
X-LIC-LOCATION:Asia/Tokyo
BEGIN:STANDARD
TZOFFSETFROM:+0900
TZOFFSETTO:+0900
TZNAME:JST
DTSTART:18871231T000000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260817T171531Z
LOCATION:Hall B5 (2)\, B Block\, Level 5
DTSTART;TZID=Asia/Tokyo:20241205T090000
DTEND;TZID=Asia/Tokyo:20241205T091400
UID:siggraphasia_SIGGRAPH Asia 2024_sess125_papers_319@linklings.com
SUMMARY:A Statistical Approach to Monte Carlo Denoising
DESCRIPTION:Hiroyuki Sakai and Christian Freude (Technical University of V
 ienna), Thomas Auzinger (Institute of Science and Technology Austria), and
  David Hahn and Michael Wimmer (Technical University of Vienna)\n\nThe sto
 chastic nature of modern Monte Carlo (MC) rendering methods inevitably pro
 duces noise in rendered images for a practical number of samples per pixel
 . The problem of denoising these images has been widely studied, with most
  recent methods relying on data-driven, pretrained neural networks. In con
 trast, in this paper we propose a statistical approach to the denoising pr
 oblem, treating each pixel as a random variable and reasoning about its di
 stribution. Considering a pixel of the noisy rendered image, we formulate 
 fast pair-wise statistical tests—based on online estimators—to decide whic
 h of the nearby pixels to exclude from the denoising filter. We show that 
 for symmetric pixel weights and normally distributed samples, the classica
 l Welch t-test is optimal in terms of mean squared error. We then show how
  to extend this result to handle non-normal distributions, using more rece
 nt confidence-interval formulations in combination with the Box-Cox transf
 ormation. Our results show that our statistical denoising approach matches
  the performance of state-of-the-art neural image denoising without having
  to resort to any computation-intensive pretraining. Furthermore, our appr
 oach easily generalizes to other quantities besides pixel intensity, which
  we demonstrate by showing additional applications to Russian roulette pat
 h termination and multiple importance sampling.\n\nRegistration Category: 
 Full Access, Full Access Supporter\n\nLanguage Format: English Language\n\
 nSession Chair: Wenzel Jakob (École Polytechnique Féderale de Lausanne (EP
 FL), NVIDIA)\n\n
URL:https://asia.siggraph.org/2024/program/?id=papers_319&sess=sess125
END:VEVENT
END:VCALENDAR