BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Below the Surface of the Non-Local Bayesian Image Denoising Method
  - Mila Nikolova (speaker) and co-author Pablo Arias from ENS Cachan
DTSTART:20170717T120000Z
DTEND:20170717T130000Z
UID:TALK73431@talks.cam.ac.uk
CONTACT:Carola-Bibiane Schoenlieb
DESCRIPTION:The non-local Bayesian (NLB) patch-based approach of Lebrun\, 
 Buades\, and Morel [1] is considered as a state-of-the-art method for the 
 restoration of (color) images corrupted by white Gaussian noise. It gave r
 ise to numerous ramifications like e.g.\, possible improvements\, process
 ing of various data sets and video. This article is the first attempt to 
 analyse the method in depth in order to understand the main phenomena unde
 rlying its effectiveness. Our analysis\, corroborated by numerical tests\,
  shows several unexpected facts. In a variational setting\, the first-ste
 p Bayesian approach to learn the prior for patches is equivalent to a pseu
 do-Tikhonov regularisation where the regularisation parameters can be posi
 tive or negative. Practically very good results in this step are mainly du
 e to the aggregation stage - whose importance needs to be re-evaluated.\n\
 n\nThis is joint work with Pablo Arias.\n\nReference\n[1] Lebrun\, M.\, Bu
 ades\, A.\, Morel\, J.M.: A nonlocal Bayesian image denoising algorithm. S
 IAM J. Imaging Sci.6(3)\, 1665-1688 (2013)
LOCATION:MR 14\, Centre for Mathematical Sciences
END:VEVENT
END:VCALENDAR