BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Asia/Hong_Kong
X-LIC-LOCATION:Asia/Hong_Kong
BEGIN:STANDARD
TZOFFSETFROM:+0800
TZOFFSETTO:+0800
TZNAME:HKT
DTSTART:19911015T033000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20251218T030657Z
LOCATION:Meeting Room S423+S424\, Level 4
DTSTART;TZID=Asia/Hong_Kong:20251216T134200
DTEND;TZID=Asia/Hong_Kong:20251216T135300
UID:siggraphasia_SIGGRAPH Asia 2025_sess121_papers_2056@linklings.com
SUMMARY:ELAD: Blind Face Restoration using Expectation-based Likelihood Ap
 proximation and Diffusion Prior
DESCRIPTION:Sean Man, Guy Ohayon, Ron Raphaeli, and Matan Kleiner (Technio
 n – Israel Institute of Technology) and Michael Elad (Technion – Israel In
 stitute of Technology, NVIDIA)\n\nBlind Face Restoration (BFR) aims to rec
 over face images suffering from unknown degradations. A recent approach to
  solve BFR is via plug-and-play methods for image restoration, which combi
 ne a likelihood function with pre-trained diffusion models as priors. Howe
 ver, as the likelihood is inherently unknown in BFR, existing methods rely
  instead on heuristic constraints. This leads to suboptimal distortion and
  identity preservation metrics. We introduce Expectation-based Likelihood 
 Approximation with Diffusion prior (ELAD), a novel plug-and-play approach 
 that explicitly models the likelihood function for BFR. ELAD estimates the
  first and second moments of the likelihood distribution by employing a De
 gradation Estimator to predict the degradation sequence from the input. Th
 is enables principled Bayesian inference without requiring end-to-end trai
 ning. Our method achieves state-of-the-art distortion and identity preserv
 ation results compared to existing plug-and-play BFR techniques, while mai
 ntaining competitive perceptual quality. As we show, while being plug-and-
 play, our method still rivals end-to-end trained BFR models.\n\nRegistrati
 on Category: Full Access, Full Access Supporter\n\nSession Chair: Paul Deb
 evec (Eyeline, USC Institute for Creative Technologies (ICT))\n\n
END:VEVENT
END:VCALENDAR