Skip to main navigation Skip to Content

Computer Science

University of Toronto
  • U of T Portal
  • Student Support
  • Contact
  • About
    • Why Study CS at U of T
    • Career Options
    • History of DCS
    • Giving to DCS
    • Computer Science at UofT Mississauga
    • Computer Science at UofT Scarborough
    • Contact
    • Employment Opportunities for Faculty/Lecturers
    • How to Find Us
  • Undergraduate
    • Prospective Undergraduates
    • Current Undergraduates
  • Graduate
    • Prospective Graduate students
    • Current Graduate students
  • Research
    • Research Areas
    • Partner with us
  • People
    • Faculty
    • Staff
    • In Memoriam
    • Alumni and Friends
    • Honours & Awards
    • Women in Computer Science
    • Graduate Student Society
    • Undergraduate Student Union
    • Undergraduate Artificial Intelligence Group
    • Undergraduate Theory Group
  • News & Events
    • News
    • Events
    • @DCS Update
    • Alumni
    • Donate
You are viewing: > Home > News & Events > Events > AI: Machine Learning Seminar - August 1
  • About
  • Undergraduate
  • Graduate
  • Research
  • People
  • News & Events

AI: Machine Learning Seminar - August 1

Event date: Monday, August 01, 2011, at 11:10 AM
Location: PT 290C

Title: Two Topics in MAP-MRF Inference

Speaker: Vladimir Kolmogorov, University College London

Abstract:
Algorithms for MAP-MRF inference (computing maximum a posteriori configuration in a Markov Random Field) are of fundamental importance for many computer vision problems. I will talk about two such algorithms.

The first one, TRW-S, is a message passing technique for general MRFs with unary and pairwise terms, which is a variation of the tree-reweighted message algorithm by Wainwright et al.. Unlike Wainwright's techniques, TRW-S has certain convergence guarantees, and also performs better in practice.

Then I will consider the problem of computing correspondences between sparse image features related by an unknown non-rigid transformation and corrupted by clutter and occlusions. We formulate it as an instance of the graph matching optimisation problem, which is NP-hard in general. To tackle it, we use the dual decomposition approach with a particular choice of subproblems. Our method was able to find the global minimum within a minute in the majority of our examples. (These examples involved about 1000 potential correspondences.) This part of the talk is a joint work with L. Torresani and C. Rother.

All rights reserved copyright Computer Science, University of Toronto | Site Map