Shape-Motion Based Athlete Tracking for Multilevel Action Recognition (original) (raw)

Abstract

An automatic human shape-motion analysis method based on a fusion architecture is proposed for human action recognition in videos. Robust shape-motion features are extracted from human points detection and tracking. The features are combined within the Transferable Belief Model (TBM) framework for action recognition. The TBM-based modelling and fusion process allows to take into account imprecision, uncertainty and conflict inherent to the features. Action recognition is performed by a multilevel analysis. The sequencing is exploited for feedback information extraction in order to improve tracking results. The system is tested on real videos of athletics meetings to recognize four types of jumps: high jump, pole vault, triple jump and long jump.

Preview

Unable to display preview. Download preview PDF.

Similar content being viewed by others

References

  1. Aggarwal, J., Park, S.: Human motion: Modeling and recognition of actions and interactions. In: 3DPVT 2004, pp. 640–647 (2004)
    Google Scholar
  2. Wang, L., Hu, W., Tan, T.: Recent developments in human motion analysis. PR 36(3), 585–601 (2003)
    Google Scholar
  3. Panagiotakis, C., Tziritas, G.: Recognition and tracking of the members of a moving human body. In: Perales, F.J., Draper, B.A. (eds.) AMDO 2004. LNCS, vol. 3179, pp. 86–98. Springer, Heidelberg (2004)
    Chapter Google Scholar
  4. Odobez, J.M., Bouthemy, P.: Robust multiresolution estimation of parametric motion models. J. of Vis. Comm. and Image R. 6(4), 348–365 (1995)
    Article Google Scholar
  5. Luo, Y., Wu, T.D., Hwang, J.N.: Object-based analysis and interpretation of human motion in sports video sequences by dynamic bayesian networks. CVIU 92, 196–216 (2003)
    Google Scholar
  6. Rombaut, M., Jarkass, I., Denœux, T.: State recognition in discrete dynamical systems using petri nets and evidence theory. In: Hunter, A., Parsons, S. (eds.) ECSQARU 1999. LNCS, vol. 1638, p. 352. Springer, Heidelberg (1999)
    Chapter Google Scholar
  7. Ramasso, E., Pellerin, D., Panagiotakis, C., Rombaut, M., Tziritas, G., Lim, W.: Spatio-temporal information fusion for human action recognition in videos. In: 13th European Signal Processing Conf. (2005)
    Google Scholar
  8. Smets, P., Kennes, R.: The Transferable Belief Model. Artificial Intelligence 66(2), 191–234 (1994)
    Article MATH MathSciNet Google Scholar
  9. Ramasso, E., Rombaut, M., Pellerin, D.: A temporal belief filter improving human action recognition in videos. In: ICASSP (to appear, 2006)
    Google Scholar

Download references

Author information

Authors and Affiliations

  1. Department of Computer Science, University of Crete, P.O. Box 2208, Heraklion, Greece
    Costas Panagiotakis & Georgios Tziritas
  2. Laboratoire des Images et des Signaux, 46 avenue Félix Viallet, 38031, Grenoble, France
    Emmanuel Ramasso, Michèle Rombaut & Denis Pellerin

Authors

  1. Costas Panagiotakis
  2. Emmanuel Ramasso
  3. Georgios Tziritas
  4. Michèle Rombaut
  5. Denis Pellerin

Editor information

Editors and Affiliations

  1. Unitat de Gràfics i Visió per Ordinador Departament de Ciències Matemàtiques i Informàtica, Universitat de les Illes Balears Edifici Anselm Turmeda, Ctra. de Valldemossa km 7,5, 07122, Palma de Mallorca, Spain
    Francisco J. Perales
  2. School of Informatics, University of Edinburgh, James Clerk Maxwell Building, The King’s Buildings, Mayfield Road, EH9 3JZ, Edinburgh, UK
    Robert B. Fisher

Rights and permissions

© 2006 Springer-Verlag Berlin Heidelberg

About this paper

Cite this paper

Panagiotakis, C., Ramasso, E., Tziritas, G., Rombaut, M., Pellerin, D. (2006). Shape-Motion Based Athlete Tracking for Multilevel Action Recognition. In: Perales, F.J., Fisher, R.B. (eds) Articulated Motion and Deformable Objects. AMDO 2006. Lecture Notes in Computer Science, vol 4069. Springer, Berlin, Heidelberg. https://doi.org/10.1007/11789239\_40

Download citation

Keywords

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.

Publish with us