Enhancing Particle Filters Using Local Likelihood Sampling

Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3021)


Particle filters provide a means to track the state of an object even when the dynamics and the observations are non-linear/non-Gaussian. However, they can be very inefficient when the observation noise is low as compared to the system noise, as it is often the case in visual tracking applications. In this paper we propose a new two-stage sampling procedure to boost the performance of particle filters under this condition. We provide conditions under which the new procedure is proven to reduce the variance of the weights. Synthetic and real-world visual tracking experiments are used to confirm the validity of the theoretical analysis.


Importance Sampling Visual Tracking License Plate Proposal Density Test Video Sequence 
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.
    Pitt, M.K., Shephard, N.: Filtering via simulation: Auxiliary particle filter. Journal of the American Statistical Association 94, 590–599 (1999)zbMATHCrossRefMathSciNetGoogle Scholar
  2. 2.
    Fox, D., Thrun, S., Burgard, W., Dellaert, F.: Particle filters for mobile robot localization. In: Doucet, A., de Freitas, N., Gordon, N. (eds.) Sequential Monte Carlo Methods in Practice, Springer, New York (2001)Google Scholar
  3. 3.
    Torma, P., Szepesvári, C.: LS-N-IPS: an improvement of particle filters by means of local search. In: Proc. Non-Linear Control Systems(NOLCOS 2001), St. Petersburg, Russia (2001)Google Scholar
  4. 4.
    Liu, J.S., Chen, R.: Sequential Monte Carlo methods for dynamic systems. Journal of the American Statistical Association 93, 1032–1044 (1998)zbMATHCrossRefMathSciNetGoogle Scholar
  5. 5.
    Liu, J.S.: Metropolized independent sampling. Statistics and Computing 6, 113–119 (1996)CrossRefGoogle Scholar
  6. 6.
    Kong, A.: A note on importance sampling using renormalized weights. Technical report (1992)Google Scholar
  7. 7.
    Rubin, D.: Using the SIR algorithm to simulate posterior distributions. In: Bernardo, J.M., DeGroot, M.H. (eds.) Bayesian Statistics 3, pp. 395–402. Oxford University Press, Oxford (1988)Google Scholar
  8. 8.
    Isard, M., Blake, A.: CONDENSATION – conditional density propagation for visual tracking. International Journal Computer Vision 29, 5–28 (1998)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2004

Authors and Affiliations

  1. 1.Tateyama Ltd.BudapestHungary
  2. 2.Computer Automation Institute of the Hungarian Academy of SciencesBudapestHungary

Personalised recommendations