BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Enforcing topological constraints in energy-based image segmentati
 on - Christoph Lampert\, IST Austria
DTSTART:20110331T120000Z
DTEND:20110331T130000Z
UID:TALK30515@talks.cam.ac.uk
CONTACT:Microsoft Research Cambridge Talks Admins
DESCRIPTION:Current techniques for energy-based image segmentation are not
  well suited to the incorporation of topological information about image r
 egions\, such as their connectedness or holefreeness. Even in the simplest
  conceivable cases\, the inclusion of topological side constraints turns t
 he necessary energy minimization steps into NP hard optimization problems.
 \nIn my talk I will present a different approach to enforce topological pr
 operties in energy-based image segmentation. Instead of formulating side c
 onstraints one searches a minimal pertubation of the unary potentials such
  that unconstrained optimization leads to a segmentation with the intended
  properties. When measuring similarity by the L^1 norm\, this setup is equ
 ivalent to the constraint-based setup (and therefore again NP-hard). Howev
 er\, when using an L^infinity-norm\, the problem becomes efficiently solva
 ble using tools from computational topology.\nBased on this observation\, 
 I will present an efficient iterative segmentation algorithm that allows i
 mage segmentation with specified topological properties even for large ima
 ges. The algorithm can also easily be extended\, e.g. to recent models wit
 h higher order potentials\, because incorporating topological constraints 
 through modified unary potentials makes the method independent of the actu
 al algorithm used for energy minimization.
LOCATION:Small public lecture room\, Microsoft Research Ltd\, 7 J J Thomso
 n Avenue (Off Madingley Road)\, Cambridge
END:VEVENT
END:VCALENDAR