Advertisement

Window Annealing over Square Lattice Markov Random Field

  • Ho Yub Jung
  • Kyoung Mu Lee
  • Sang Uk Lee
Part of the Lecture Notes in Computer Science book series (LNCS, volume 5303)

Abstract

Monte Carlo methods and their subsequent simulated annealing are able to minimize general energy functions. However, the slow convergence of simulated annealing compared with more recent deterministic algorithms such as graph cuts and belief propagation hinders its popularity over the large dimensional Markov Random Field (MRF). In this paper, we propose a new efficient sampling-based optimization algorithm called WA (Window Annealing) over squared lattice MRF, in which cluster sampling and annealing concepts are combined together. Unlike the conventional annealing process in which only the temperature variable is scheduled, we design a series of artificial ”guiding” (auxiliary) probability distributions based on the general sequential Monte Carlo framework. These auxiliary distributions lead to the maximum a posteriori (MAP) state by scheduling both the temperature and the proposed maximum size of the windows (rectangular cluster) variable. This new annealing scheme greatly enhances the mixing rate and consequently reduces convergence time. Moreover, by adopting the integral image technique for computation of the proposal probability of a sampled window, we can achieve a dramatic reduction in overall computations. The proposed WA is compared with several existing Monte Carlo based optimization techniques as well as state-of-the-art deterministic methods including Graph Cut (GC) and sequential tree re-weighted belief propagation (TRW-S) in the pairwise MRF stereo problem. The experimental results demonstrate that the proposed WA method is comparable with GC in both speed and obtained energy level.

Keywords

Simulated Annealing Markov Random Field Global Constraint Sequential Monte Carlo Detailed Balance Condition 
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.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Sun, J., Zheng, N.N., Shum, H.Y.: Stereo matching using belief propagation. PAMI 25 (2003)Google Scholar
  2. 2.
    Boykov, Y., Veksler, O., Zabih, R.: Fast approximate energy minimization via graph cuts. PAMI 23 (2001)Google Scholar
  3. 3.
    Potetz, B.: Efficient belief propagation for vision using linear constraint nodes. CVPR (2007)Google Scholar
  4. 4.
    Rother, C., Kolmogorov, V., Minka, T., Blake, A.: Cosegmenation of image pairs by histogram matching- incorporating a global constraint into mrfs. CVPR (2006)Google Scholar
  5. 5.
    Rother, C., Kolmogorov, V., Lepistsky, V., Szummer, M.: Optimizing binary mrfs via extended roof duality. CVPR (2007)Google Scholar
  6. 6.
    Kohli, P., Mudigonda, P., Torr, P.: p 3 and beyond: Solving energies with higher order cliques. CVPR (2007)Google Scholar
  7. 7.
    Lan, X., Roth, S., Huttenlocher, D., Black, M.J.: Efficient belief propagation with learned higher-order markov random fields. In: ECCV (2006)Google Scholar
  8. 8.
    Barbu, A., Zhu, S.C.: Generalizing swendsen-wang cut to sampling arbitrary posterior probabilities. PAMI 27 (2005)Google Scholar
  9. 9.
    Geman, S., Geman, D.: Stochastic relaxation, gibbs distributions, and the bayesian restoration of images. PAMI 6 (1984)Google Scholar
  10. 10.
    Tu, Z., Zhu, S.C.: Image segmentation by data-driven markov chain monte carlo. PAMI 24 (May 2002)Google Scholar
  11. 11.
    Zhu, S.C., Liu, X.W., Wu, Y.N.: Exploring texture ensembles by efficent markov chain monte carlo: Toward a trichromacy theory of texture. PAMI 22(6) (2000)Google Scholar
  12. 12.
    Liu, J.S.: Monte carlo strategies in scientific computing. Springer, Heidelberg (2001)zbMATHGoogle Scholar
  13. 13.
    Moral, P.D., Doucet, A., Jasra, A.: Sequential monte carlo for bayesian computation. Bayesian Statics (2006)Google Scholar
  14. 14.
    Metropolis, N., Rosenbluth, A.W., Rosenbluth, M.N.: Equations of state calculations by fast computing machines. Journal of Chemical Physics 21 (1953)Google Scholar
  15. 15.
    Hasting, W.K.: Monte carlo sampling methods using markov chains and their applications. Biometrika 57 (1970)Google Scholar
  16. 16.
    Kirkpatrick, S., Gelatt, C.D., Vechi, M.P.: Optimization by simulated annealing. Science 220 (1983)Google Scholar
  17. 17.
    Jasra, A., Stephens, D.A., Holmes, C.C.: On population-based simulation for static inference. Statistics and Computing 17 (2007)Google Scholar
  18. 18.
    Neal, R.M.: Annealed importance sampling. Technical Report 9805, University of Toronto (1998)Google Scholar
  19. 19.
    Andrieu, C., Freitas, N., Doucet, A., Jordan, M.: An introduction to mcmc for machine learning. Machine Learning 50 (2003)Google Scholar
  20. 20.
    Crow, F.: Summed-area tables for texture mapping. SIGGRAPH (1984)Google Scholar
  21. 21.
    Tappen, M., Freeman, W.: Comparison of graph cuts with belief propagation for stereo, using identical mrf parameters. In: ICCV (2003)Google Scholar
  22. 22.
    Szeliski, R., Zabih, R., Scharstein, D., Veksler, O., Kolmogorov, V., Agarwala, A., Tappen, M., Rother, C.: A comparative study of energy minimization methods for markov random fields. In: ECCV (2006)Google Scholar
  23. 23.
    Scharstein, D., Szeliski, R.: A taxonomy and evaluation of dense two-frame stereo correspondence algorithms. IJCV (2002)Google Scholar
  24. 24.
  25. 25.
    Scharstein, D., Szeliski, R.: High-accuracy stereo depth maps using structured light. CVPR (2003)Google Scholar
  26. 26.
    Kolmogorov, V., Zabih, R.: What energy functions can be minimized via graph cuts? PAMI 26 (2004)Google Scholar
  27. 27.
    Boykov, Y., Kolmogorov, V.: An experimental comparison of min-cut/max-flow algorithms for energy minimization in vision. PAMI 26 (2004)Google Scholar
  28. 28.
    Wainwright, M.J., Jaakkola, T.S., Willsky, A.S.: Map estimation via agreement on trees: Message-passing and linear-programming approaches. IEEE Trans. Information Theory 51(11) (2005)Google Scholar
  29. 29.
    Kolmogorov, V.: Convergent tree-reweighted message passing for energy minimization. PAMI 28 (2006)Google Scholar
  30. 30.
    Sudderth, E.B., Ihler, A.T., Freeman, W.T., Willsky, A.S.: Nonparametric belief propagation. MIT LIDS Technical Report P-2551 (2002)Google Scholar
  31. 31.
    Birchfield, S., Tomasi, C.: A pixel dissimilarity measure that is insensiitive to image samplin. PAMI 20 (1998)Google Scholar
  32. 32.
    Rother, C., Kolmogorov, V., Minka, T., Blake, A.: Cosegmentation of image pairs by histogram matchin- incorporating a global constraint into mrfs. CVPR (2006)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2008

Authors and Affiliations

  • Ho Yub Jung
    • 1
  • Kyoung Mu Lee
    • 1
  • Sang Uk Lee
    • 1
  1. 1.Department of EECS, ASRISeoul National UniversitySeoulKorea

Personalised recommendations