Particle-Based Belief Propagation for Structure from Motion and Dense Stereo Vision with Unknown Camera Constraints

  • Hoang Trinh
  • David McAllester
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4931)


In this paper, we present a specific use of the Particle-based Belief Propagation (PBP) algorithm as an approximation scheme for the joint distribution over many random variables with very large or continuous domains. After formulating the problem to be solved as a probabilistic graphical model, we show that by running loopy Belief Propagation on the whole graph, in combination with an MCMC method such as Metropolis-Hastings sampling at each node, we can approximately estimate the posterior distribution of each random variable over the state space. We describe in details the application of PBP algorithm to the problem of sparse Structure from Motion and the dense Stereo Vision with unknown camera constraints. Experimental results from both cases are demonstrated. An experiment with a synthetic structure from motion arrangement shows that its accuracy is comparable with the state-of-the-art while allowing estimates of state uncertainty in the form of an approximate posterior density function.


Belief Propagation Particle filter Structure from Motion Dense Stereo Vision 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Pearl, J.: Probabilistic Reasoning in Intelligent Systems. Morgan Kaufman, San Mateo (1988)Google Scholar
  2. 2.
    Freeman, W.T., Pasztor, E.C., Carmichael, O.T.: Learning low-level vision. IJCV 40(1), 25–47 (2000)zbMATHCrossRefGoogle Scholar
  3. 3.
    Coughlan, J.M., Ferreira, S.J.: Finding deformable shapes using loopy belief propagation. ECCV 7, 453–468 (2002)Google Scholar
  4. 4.
    Coughlan, J.M., Shen, H.: Shape matching with belief propagation: Using dynamic quantization to accomodate occlusion and clutter. In: CVPR Workshop on Generative Model Based Vision (2004)Google Scholar
  5. 5.
    Arulampalam, M.S., et al.: A tutorial on particle filters for online nonlinear/non-Gaussian Bayesian tracking. IEEE Trans. SP 50(2), 174–188 (2002)CrossRefGoogle Scholar
  6. 6.
    Koller, D., Lerner, U., Angelov, D.: A general algorithm for approximate inference and its application to hybrid Bayes nets. UAI 15, 324–333 (1999)Google Scholar
  7. 7.
    Sudderth, E.B., et al.: Nonparametric belief propagation. In: CVPR (2003)Google Scholar
  8. 8.
    Isard, M.: PAMPAS: Real–valued graphical models for computer vision. In: CVPR (2003)Google Scholar
  9. 9.
    Neal, R.M., Beal, M.J., Roweis, S.T.: Inferring state sequences for non-linear systems with embedded hidden Markov models. In: NIPS, vol.16 (2003)Google Scholar
  10. 10.
    Thrun, S., Fox, D., Burgard, W.: Monte carlo localization with mixture proposal distribution. In: AAAI, pp. 859–865 (2000)Google Scholar
  11. 11.
    van der Merwe, R., et al.: The unscented particle filter. In: NIPS, vol. 13 (2001)Google Scholar
  12. 12.
    Doucet, A., de Freitas, N., Gordon, N. (eds.): Sequential Monte Carlo Methods in Practice. Springer, New York (2001)zbMATHGoogle Scholar
  13. 13.
    Godsill, S., Clapp, T.: Improvement strategies for Monte Carlo particle filters. In: Doucet, J.F.G.D.F.A., Gordon, N.J. (eds.) Sequential Monte Carlo Methods in Practice, Springer, New York (2001)Google Scholar
  14. 14.
    Khan, Z., Balch, T., Dellaert, F.: MCMC-based particle filtering for tracking a variable number of interacting targets. IEEE Trans. PAMI, 1805–1918 (2005)Google Scholar
  15. 15.
    Clifford, P.: Markov random fields in statistics. In: Grimmett, G.R., Welsh, D.J.A. (eds.) Disorder in Physical Systems, pp. 19–32. Oxford University Press, Oxford (1990)Google Scholar
  16. 16.
    Triggs, B., et al.: Bundle adjustment - a modern synthesis. In: Vision Algorithms (1999)Google Scholar
  17. 17.
    Hartley, R.I., Zisserman, A.: Multiple View Geometry in Computer Vision, 2nd edn. Cambridge University Press, Cambridge (2004)zbMATHGoogle Scholar
  18. 18.
    Kaucic, R., Hartley, R., Dano, N.: Plane-based projective reconstruction. In: ICCV 2001. Proceedings of the Ninth IEEE International Conference on Computer Vision. IEEE Computer Society Press, Los Alamitos (2001)Google Scholar
  19. 19.
    Snavely, N., Seitz, S.M., Szeliski, R.: Photo tourism: Exploring photo collections in 3D. ACM Transactions on Graphics 25(3), 835–846 (2006)CrossRefGoogle Scholar
  20. 20.
    Lowe, D.G.: Object recognition from local scale-invariant features. In: Proc. of the International Conference on Computer Vision, Corfu., pp. 1150–1157 (1999)Google Scholar
  21. 21.
    Lourakis, M., Argyros, A.: The design and implementation of a generic sparse bundle adjustment software package based on the levenberg-marquardt algorithm. In: Technical Report 340, Institute of Computer Science - FORTH, Heraklion, Crete, Greece (2004),
  22. 22.
    Ihler, A.T., et al.: Nonparametric belief propagation for self-calibration in sensor networks. IEEE Trans. Jsac., 809–819 (2005)Google Scholar
  23. 23.
    Scharstein, D., Szeliski, R., Zabih, R.: A taxonomy and evaluation of dense two-frame stereo correspondence algorithms (2001)Google Scholar
  24. 24.
    Sun, J., Zheng, N.N., Shum, H.Y.: Stereo matching using belief propagation. IEEE Trans. Pattern Anal. Mach. Intell. 25(7), 787–800 (2003)CrossRefGoogle Scholar
  25. 25.
    Felzenszwalb, P.F., Huttenlocher, D.P.: Efficient belief propagation for early vision. Int. J. Comput. Vision 70(1), 41–54 (2006)CrossRefGoogle Scholar
  26. 26.
    Weiss, F.: On the optimality of solutions of the max-product belief-propagation algorithm in arbitrary graphs. IEEETIT: IEEE Transactions on Information Theory 47 (2001)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2008

Authors and Affiliations

  • Hoang Trinh
    • 1
  • David McAllester
    • 1
  1. 1.Toyota Technological Institute at ChicagoChicago Illinois 

Personalised recommendations