BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Denoising Geometric Image Features - Stacey Levine (Duquesne Unive
 rsity)
DTSTART:20171031T090000Z
DTEND:20171031T095000Z
UID:TALK94105@talks.cam.ac.uk
CONTACT:INI IT
DESCRIPTION:Given a noisy image\, it can sometimes be more productive to d
 enoise a transformed version of the image rather than process the image da
 ta directly. In this talk we will discuss several novel frameworks for ima
 ge denoising\, including one that involves smoothing the noisy image&rsquo
 \;s level line curvature and another that regularizes the components of th
 e noisy image in a moving frame that encodes its local geometry. Both fram
 eworks satisfy some nice unexpected properties that provide justification 
 for this framework. Experiments confirm an improvement over the usual deno
 ising paradigm in terms of both PSNR and SSIM. Moreover\, this approach pr
 ovides a mechanism for preserving geometry in solutions of sparse patch ba
 sed models that typically exploit self similarity. This is joint work with
  Thomas Batard\, Marcelo Bertalmio\, and Gabriela Ghimpeteanu.
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR