Beyond Loose LP-Relaxations: Optimizing MRFs by Repairing Cycles

  • Nikos Komodakis
  • Nikos Paragios
Part of the Lecture Notes in Computer Science book series (LNCS, volume 5304)


This paper presents a new MRF optimization algorithm, which is derived from Linear Programming and manages to go beyond current state-of-the-art techniques (such as those based on graph-cuts or belief propagation). It does so by relying on a much tighter class of LP-relaxations, called cycle-relaxations. With the help of this class of relaxations, our algorithm tries to deal with a difficulty lying at the heart of MRF optimization: the existence of inconsistent cycles. To this end, it uses an operation called cycle-repairing. The goal of that operation is to fix any inconsistent cycles that may appear during optimization, instead of simply ignoring them as usually done up to now. The more the repaired cycles, the tighter the underlying LP relaxation becomes. As a result of this procedure, our algorithm is capable of providing almost optimal solutions even for very general MRFs with arbitrary potentials. Experimental results verify its effectiveness on difficult MRF problems, as well as its better performance compared to the state of the art.


Dual Solution Anchor Node Linear Programming Relaxation Tight Link Dual Objective 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


  1. 1.
    Komodakis, N., Tziritas, G.: Approximate labeling via graph-cuts based on linear programming. In: PAMI (2007)Google Scholar
  2. 2.
    Komodakis, N., Paragios, N., Tziritas, G.: MRF Optimization via Dual Decomposition: Message-passing revisited. In: ICCV (2007)Google Scholar
  3. 3.
    Wainwright, M.J., Jaakkola, T.S., Willsky, A.S.: Map estimation via agreement on (hyper)trees: Message-passing and linear programming. IEEE Trans. on Information Theory (2005)Google Scholar
  4. 4.
    Kolmogorov, V.: Convergent Tree-reweighted Message Passing for Energy Minimization. PAMI 28(10) (2006)Google Scholar
  5. 5.
    Komodakis, N., Tziritas, G., Paragios, N.: Fast Approximately Optimal Solutions for Single and Dynamic MRFs. In: CVPR (2007)Google Scholar
  6. 6.
    Werner, T.: A Linear Programming Approach to Max-Sum Problem: A Review. PAMI 29(7) (2007)Google Scholar
  7. 7.
    Yanover, C., Meltzer, T., Weiss, Y.: Linear Programming Relaxations and Belief Propagation – An Empirical Study. JMLR 7 (2006)Google Scholar
  8. 8.
    Sontag, D., Jaakkola, T.: New outer bounds on the marginal polytope. In: NIPS (2008)Google Scholar
  9. 9.
    Werner, T.: High-arity interactions, polyhedral relaxations, and cutting plane algorithm for soft constraint optimisation (MAP-MRF). In: CVPR (2008)Google Scholar
  10. 10.
    Sontag, D., Meltzer, T., Globerson, A., Jaakkola, T., Weiss, Y.: Tightening LP relaxations for MAP using message passing. In: UAI (2008)Google Scholar
  11. 11.
    Kumar, M.P., Kolmogorov, V., Torr, P.H.S.: An analysis of convex relaxations for MAP estimation. In: NIPS (2007)Google Scholar
  12. 12.
    Komodakis, N., Paragios, N.: Beyond loose LP-relaxations for MRF optimization. Technical report, Ecole Centrale Paris (2008)Google Scholar
  13. 13.
    Szeliski, R., Zabih, R., et al.: A Comparative Study of Energy Minimization Methods for Markov Random Fields. In: Leonardis, A., Bischof, H., Pinz, A. (eds.) ECCV 2006. LNCS, vol. 3952, pp. 16–29. Springer, Heidelberg (2006)CrossRefGoogle Scholar
  14. 14.
    Boykov, Y., Veksler, O., Zabih, R.: Fast Approximate Energy Minimization via Graph Cuts. In: PAMI (November 2001)Google Scholar
  15. 15.
    Raj, A., Singh, G., Zabih, R.: MRF’s for MRI’s: Bayesian Reconstruction of MR Images via Graph Cuts. In: CVPR (2006)Google Scholar
  16. 16.
    Rother, C., Kolmogorov, V., Lempitsky, V., Szummer, M.: Optimizing binary MRFs via extended roof duality. In: CVPR (2007)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2008

Authors and Affiliations

  • Nikos Komodakis
    • 1
  • Nikos Paragios
    • 2
  1. 1.University of CreteGreece
  2. 2.Ecole Centrale de ParisFrance

Personalised recommendations