Probabilistic Spatio-temporal 2D-Model for Pedestrian Motion Analysis in Monocular Sequences

  • Grégory Rogez
  • Carlos Orrite
  • Jesús Martínez
  • J. Elías Herrero
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4069)


This paper addresses the problem of probabilistic modelling of human motion by combining several 2D views. This method takes advantage of 3D information avoiding the use of a complex 3D model. Considering that the main disadvantage of 2D models is their restriction to the camera angle, a solution to this limitation is proposed in this paper. A multi-view Gaussian Mixture Model (GMM) is therefore fitted to a feature space made of Shapes and Stick figures manually labelled. Temporal and spatial constraints are considered to build a probabilistic transition matrix. During the fitting, this matrix limits the feature space only to the most probable models from the GMM. Preliminary results have demonstrated the ability of this approach to adequately estimate postures independently of the direction of motion during the sequence.


Gaussian Mixture Model Gait Cycle Probabilistic Transition Matrix Spatial Constraint Temporal Cluster 
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.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Wang, L., Hu, W., Tan, T.: Recent developments in human motion analysis. Pattern Recognition 36, 585–601 (2003)CrossRefGoogle Scholar
  2. 2.
    Kakadiaris, I.A., Metaxas, D.N.: Model-based estimation of 3d human motion. IEEE Trans. Pattern Anal. Mach. Intell. 22, 1453–1459 (2000)CrossRefGoogle Scholar
  3. 3.
    Sidenbladh, H., Black, M.J., Sigal, L.: Implicit probabilistic models of human motion for synthesis and tracking. In: Heyden, A., Sparr, G., Nielsen, M., Johansen, P. (eds.) ECCV 2002. LNCS, vol. 2350, pp. 784–800. Springer, Heidelberg (2002)CrossRefGoogle Scholar
  4. 4.
    Baumberg, A., Hogg, D.: Learning flexible models from image sequences. In: Eklundh, J.-O. (ed.) ECCV 1994. LNCS, vol. 800, pp. 299–308. Springer, Heidelberg (1994)Google Scholar
  5. 5.
    Sminchisescu, C., Triggs, B.: Kinematic jump processes for monocular 3d human tracking. In: CVPR (1), pp. 69–76 (2003)Google Scholar
  6. 6.
    Bowden, R., Mitchell, T.A., Sarhadi, M.: Reconstructing 3d pose and motion from a single camera view. In: BMVC (1998)Google Scholar
  7. 7.
    Gross, R., Shi, J.: The cmu motion of body (mobo) database, Robotics Institute, Carnegie Mellon University, Pittsburgh, PA (2001)Google Scholar
  8. 8.
    Cootes, T., Taylor, C.: A mixture model for representing shape variation (1997)Google Scholar
  9. 9.
    Ponsa, D., Roca, F.X.: A novel approach to generate multiple shape models for tracking applications. In: Perales, F.J., Hancock, E.R. (eds.) AMDO 2002. LNCS, vol. 2492, pp. 80–91. Springer, Heidelberg (2002)CrossRefGoogle Scholar
  10. 10.
    Bowden, R., Sarhadi, M.: Building temporal models for gesture recognition. In: BMVC (2000)Google Scholar
  11. 11.
    Heap, T., Hogg, D.: Wormholes in shape space: Tracking through discontinuous changes in shape. In: ICCV, pp. 344–349 (1998)Google Scholar
  12. 12.
    Inman, V.T., Ralston, H.J., Todd, F.: Human Walking. Williams and Wilkins, Baltimore (1981)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Grégory Rogez
    • 1
  • Carlos Orrite
    • 1
  • Jesús Martínez
    • 1
  • J. Elías Herrero
    • 1
  1. 1.CVLab, Aragon Institute for Engineering ResearchUniversity of ZaragozaSpain

Personalised recommendations