Student-t Mixture Filter for Robust, Real-Time Visual Tracking

  • James Loxam
  • Tom Drummond
Part of the Lecture Notes in Computer Science book series (LNCS, volume 5304)


Filtering is a key problem in modern information theory; from a series of noisy measurement, one would like to estimate the state of some system. A number of solutions exist in the literature, such as the Kalman filter or the various particle and hybrid filters, but each has its drawbacks.

In this paper, a filter is introduced based on a mixture of Student-t modes for all distributions, eliminating the need for arbitrary decisions when treating outliers and providing robust real-time operation in a true Bayesian manner.


Posterior Distribution Heavy Tail Approximate Inference High Order Cumulants Data Confusion 
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.

Supplementary material

978-3-540-88690-7_28_MOESM1_ESM.avi (20.2 mb)
Supplementary material (20,655 KB)


  1. 1.
    Huber, P.: Robust Statistics. Wiley, New York (1981)CrossRefzbMATHGoogle Scholar
  2. 2.
    Kalman, R.: A new approach to linear filtering and prediction problems. Transactions of the ASME–Journal of Basic Engineering 82, 35–45 (1960)CrossRefGoogle Scholar
  3. 3.
    Fischler, M., Bolles, R.: Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography. Commun. ACM 24(6), 381–395 (1981)MathSciNetCrossRefGoogle Scholar
  4. 4.
    Finetti, B.D.: The bayesian approach to the rejection of outliers. In: Proc. Fourth Berkeley Symp. on Math. Statist. and Prob., vol. 1, pp. 199–210 (1961)Google Scholar
  5. 5.
    Gordon, N.: A hybrid bootstrap filter for target tracking in clutter. IEEE Transactions on Aerospace and Electronic Systems (1997)Google Scholar
  6. 6.
    Isard, M., Blake, A.: Condensation – conditional density propagation for visual tracking. International Journal of Computer Vision 29(1), 5–28 (1998)CrossRefGoogle Scholar
  7. 7.
    Gordon, N., Smith, A.: Approximate non-gaussian bayesian estimation and modal consistency. Journal of the Royal Statistical Society B 55(4), 913–918 (1993)MathSciNetzbMATHGoogle Scholar
  8. 8.
    van der Merwe, R., de Freitas, J., Doucet, A., Wan, E.: The unscented particle filter. In: Advances in Neural Information Processing Systems, vol. 13 (2001)Google Scholar
  9. 9.
    Kotecha, J.H., Djuric, P.M.: Gaussian sum particle filtering. IEEE Transactions on Signal Processing 51, 2602–2612 (2003)MathSciNetCrossRefzbMATHGoogle Scholar
  10. 10.
    Han, B., Zhu, Y., Comaniciu, D., Davis, L.: Kernel-based bayesian filtering for object tracking. In: Proc. IEEE CVPR (2005)Google Scholar
  11. 11.
    Li, S., Wang, H., Chai, T.: A t-distribution based particle filter for target tracking. In: Proc. American Control Conference, pp. 2191–2196 (2006)Google Scholar
  12. 12.
    Dawid, A.P.: Posterior expectations for large observations. Miscellanea, 664–667 (1973)Google Scholar
  13. 13.
    O’Hagan, A.: On outlier rejection phenomena in bayes inference. Journal of the Royal Statistical Society B 41(3), 358–367 (1979)MathSciNetzbMATHGoogle Scholar
  14. 14.
    West, M.: Robust sequential approximate bayesian estimation. Journal of the Royal Statistical Society B 43(2), 157–166 (1981)MathSciNetzbMATHGoogle Scholar
  15. 15.
    Sorenson, H., Alspach, D.: Recursive Bayesian estimation using Gaussian sums. Automatica 7(4), 465–479 (1971)MathSciNetCrossRefzbMATHGoogle Scholar
  16. 16.
    Meinhold, R.J., Singpurwalla, N.D.: Robustification of kalman filter models. Journal of the American Statistical Association 84, 479–486 (1989)MathSciNetCrossRefGoogle Scholar
  17. 17.
    Mardia, K.: Measures of multivariate skewness and kurtosis with applications. Biometrika 57, 519–530 (1970)MathSciNetCrossRefzbMATHGoogle Scholar
  18. 18.
    Arthur, D., Vassilvitskii, S.: k-means++: the advantages of careful seeding. In: SODA 2007. Proceedings of the eighteenth annual ACM-SIAM symposium on Discrete algorithms, Philadelphia, PA, USA, pp. 1027–1035. Society for Industrial and Applied Mathematics (2007)Google Scholar
  19. 19.
    Jammalamadaka, S.R., Rao, T.S., Terdik, G.: Higher order cumulants of random vectors, differential operators, and applications to statistical inference and time series (1991)Google Scholar
  20. 20.
    Rosten, E., Drummond, T.: Machine learning for high speed corner detection. In: 9th European Conference on Computer Vision (May 2006)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2008

Authors and Affiliations

  • James Loxam
    • 1
  • Tom Drummond
    • 1
  1. 1.Department of EngineeringUniversity of CambridgeUSA

Personalised recommendations