OBJECT TRACKING BASED ON PARTICLE FILTERING WITH MULTIPLE APPEARANCE MODELS

Nicolas Widynski, Emanuel Aldea, Séverine Dubuisson, Isabelle Bloch

2011

Abstract

In this paper, we propose a novel method to track an object whose appearance is evolving in time. The tracking procedure is performed by a particle filter algorithm in which all possible appearance models are explicitly considered using a mixture decomposition of the likelihood. Then, the component weights of this mixture are conditioned by both the state and the current observation. Moreover, the use of the current observation makes the estimation process more robust and allows handling complementary features, such as color and shape information. In the proposed approach, these estimated component weights are computed using a Support Vector Machine. Tests on a mouth tracking problem show that the multiple appearance model outperforms classical single appearance likelihood.

References

  1. Ben-Hur, A., Horn, D., Siegelmann, H., and Vapnik, V. (2002). Support vector clustering. The Journal of Machine Learning Research, 2:125-137.
  2. Brasnett, P., Mihaylova, L., Bull, D., and Canagarajah, N. (2007). Sequential Monte Carlo tracking by fusing multiple cues in video sequences. Image Vision Computing, 25(8):1217-1227.
  3. Dhillon, I., Guan, Y., and Kulis, B. (2004). Kernel kmeans: spectral clustering and normalized cuts. In ACM SIGKDD, pages 551-556.
  4. Doucet, A., De Freitas, N., and Gordon, N., editors (2001). Sequential Monte Carlo methods in practice. Springer.
  5. Erdem, E., Dubuisson, S., and Bloch, I. (2010). Particle Filter-Based Visual Tracking by Fusing Multiple Cues with Context-Sensitive Reliabilities. Technical Report 2010D002, Télécom ParisTech.
  6. Hotta, K. (2006). Adaptive Weighting of Local Classifiers by Particle Filter. In ICPR, volume 2, pages 610-613.
  7. Maggio, E., Smeraldi, F., and Cavallaro, A. (2005). Combining colour and orientation for adaptive particle filter-based tracking. In British Machine Vision Conference, pages 659-668.
  8. Mun˜ oz-Salinas, R., Aguirre, E., García-Silvente, M., and Gonzalez, A. (2008). A multiple object tracking approach that combines colour and depth information using a confidence measure. Pattern Recognition Letters, 29(10):1504-1514.
  9. Nummiaro, K., Koller-Meier, E., and Gool, L. V. (2002). Object Tracking with an Adaptive Color-Based Particle Filter. In Symposium for Pattern Recognition of the DAGM, pages 353-360.
  10. Pérez, P., Hue, C., Vermaak, J., and Gangnet, M. (2002). Color-Based Probabilistic Tracking. In ECCV, pages 661-675.
  11. Platt, J. C. (2000). Probabilistic Outputs for Support Vector Machines and Comparisons to Regularized Likelihood Methods. In Advances in Large Margin Classifiers, pages 61-74.
  12. Widynski, N., Dubuisson, S., and Bloch, I. (2010). Integration of fuzzy spatial information in tracking based on particle filtering. IEEE Transactions on Systems, Man and Cybernetics SMCB, To Appear.
  13. Wu, T.-F., Lin, C.-J., and Weng, R. C. (2004). Probability estimates for multi-class classification by pairwise coupling. Journal of Machine Learning Research, 5:975-1005.
  14. Xu, X. and Li, B. (2005). Rao-Blackwellised particle filter for tracking with application in visual surveillance. In IEEE International Workshop on Visual Surveillance and Performance Evaluation of Tracking and Surveillance, pages 17-24.
Download


Paper Citation


in Harvard Style

Widynski N., Aldea E., Dubuisson S. and Bloch I. (2011). OBJECT TRACKING BASED ON PARTICLE FILTERING WITH MULTIPLE APPEARANCE MODELS . In Proceedings of the International Conference on Computer Vision Theory and Applications - Volume 1: VISAPP, (VISIGRAPP 2011) ISBN 978-989-8425-47-8, pages 604-609. DOI: 10.5220/0003334606040609


in Bibtex Style

@conference{visapp11,
author={Nicolas Widynski and Emanuel Aldea and Séverine Dubuisson and Isabelle Bloch},
title={OBJECT TRACKING BASED ON PARTICLE FILTERING WITH MULTIPLE APPEARANCE MODELS},
booktitle={Proceedings of the International Conference on Computer Vision Theory and Applications - Volume 1: VISAPP, (VISIGRAPP 2011)},
year={2011},
pages={604-609},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0003334606040609},
isbn={978-989-8425-47-8},
}


in EndNote Style

TY - CONF
JO - Proceedings of the International Conference on Computer Vision Theory and Applications - Volume 1: VISAPP, (VISIGRAPP 2011)
TI - OBJECT TRACKING BASED ON PARTICLE FILTERING WITH MULTIPLE APPEARANCE MODELS
SN - 978-989-8425-47-8
AU - Widynski N.
AU - Aldea E.
AU - Dubuisson S.
AU - Bloch I.
PY - 2011
SP - 604
EP - 609
DO - 10.5220/0003334606040609