BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Video colorization by a variational approach - Jean-Francois Aujol
  (Université Bordeaux 1)
DTSTART:20170904T151000Z
DTEND:20170904T160000Z
UID:TALK77651@talks.cam.ac.uk
CONTACT:INI IT
DESCRIPTION:This work provides a new method to colorize gray-scale images.
  While the reverse operation is only a matter of standard\, the colorizati
 on process is an ill-posed problem that requires some priors.  In the lite
 rature two classes of approach exist.  The first  class  includes  manual 
  methods  that  needs  the  user  to  manually  add  colorson  the  image 
  to  colorize.   The  second  class  includes  exemplar-based  approachesw
 here a color image\, with a similar semantic content\, is provided as inpu
 t to the method.These two types of priors have their own advantages and dr
 awbacks.  In this  work\,  a  new  variational  framework  for  exemplar-b
 ased  colorization  is  proposed.  A non-local approach is used to find re
 levant color in the source image in order to suggest colors on the gray-sc
 ale image. The spatial coherency of the result as well as the final color 
 selection is provided by a non-convex variational framework based on a tot
 al variation. An efficient primal-dual algorithm.  In this work\, we also 
 extend the proposed exemplar-based approach to combine both exemplar-based
  and manual methods. It provides a single framework that unifies advantage
 s of both approaches. Finally\,experiments and comparisons with state-of-t
 he-art methods illustrate the efficiency<span> of our method.<br>This is j
 oint work with Fabien Pierre\, Aur&eacute\;lie Bugeau\, Nicolas Papadakis\
 , and Vinh Tong Ta.<br></span>
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR