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:20250110T023312Z
LOCATION:Hall B5 (2)\, B Block\, Level 5
DTSTART;TZID=Asia/Tokyo:20241205T091400
DTEND;TZID=Asia/Tokyo:20241205T092800
UID:siggraphasia_SIGGRAPH Asia 2024_sess125_papers_563@linklings.com
SUMMARY:Online Neural Denoising with Cross-Regression for Interactive Rend
 ering
DESCRIPTION:Technical Papers\n\nHajin Choi (Gwangju Institute of Science a
 nd Technology); Seokpyo Hong (Samsung Advanced Institute of Technology); I
 nwoo Ha (Samsung Advanced Institute of Technology, KAIST); Nahyup Kang (Sa
 msung Advanced Institute of Technology); and Bochang Moon (Gwangju Institu
 te of Science and Technology)\n\nGenerating a rendered image sequence thro
 ugh Monte Carlo ray tracing is an appealing option when one aims to accura
 tely simulate various lighting effects. Unfortunately, interactive renderi
 ng scenarios limit the allowable sample size for such sampling-based light
  transport algorithms, resulting in an unbiased but noisy image sequence. 
 Image denoising has been widely adopted as a post-sampling process to conv
 ert such noisy image sequences into biased but temporally stable ones. The
  state-of-the-art strategy for interactive image denoising involves devisi
 ng a deep neural network and training this network via supervised learning
 , i.e., optimizing the network parameters using training datasets that inc
 lude an extensive set of image pairs (noisy and ground truth images). This
  paper adopts the prevalent approach for interactive image denoising, whic
 h relies on a neural network. However, instead of supervised learning, we 
 propose a different learning strategy that trains our network parameters o
 n the fly, i.e., updating them online using runtime image sequences. To ac
 hieve our denoising objective with online\nlearning, we tailor local regre
 ssion to a cross-regression form that can guide robust training of our den
 oising neural network. We demonstrate that our denoising framework effecti
 vely reduces noise in input image sequences while robustly preserving both
  geometric and non-geometric edges, without requiring the manual effort in
 volved in preparing an external dataset.\n\nRegistration Category: Full Ac
 cess, Full Access Supporter\n\nLanguage Format: English Language\n\nSessio
 n Chair: Wenzel Jakob (École Polytechnique Fédérale de Lausanne)
URL:https://asia.siggraph.org/2024/program/?id=papers_563&sess=sess125
END:VEVENT
END:VCALENDAR