International Journal of Computer Vision

, Volume 37, Issue 2, pp 209–215 | Cite as

Statistical Models of Object Interaction

  • R.J. Morris
  • D.C. Hogg


We present a method for assessing the likelihood of a trajectory of an object through a scene consisting of a number of other objects. The closest points on the trajectory to the other objects are chosen as landmark points and at each landmark we calculate the probability of the interaction based on the speed and distance. Sequences of such probabilities are then sorted in increasing order. Finally a weighted sum of the first few elements in this weighted list is used to classify trajectories in a supervised learning framework.

behaviour modelling object interactions landmark data 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Baumberg, A. and Hogg, D.C. 1994a. An efficient method for countor tracking using active shape models. In Proceedings of the IEEE Workshop on Motion on Non-rigid and Articulated Objects, IEEE Press, pp. 194–199.Google Scholar
  2. Baumberg, A. and Hogg, D.C. 1994b. Learning flexiable models from image sequences. In Proceedings of the European Conference on Computer Vision, Vol. 1, pp. 299–308.Google Scholar
  3. Baumberg, A. and Hogg, D.C. 1996. Learning spationtemporal models from examples. Image and Vision Computing, 9 (special issue BMVC).Google Scholar
  4. Bobick, A.F. and Wilson, A.D. 1995. A state-based technique for the summarisation and recognition of gesture. In Proc. 5th Int. Conf. on Computer Vision, pp. 382–388.Google Scholar
  5. Bookstein, F.L. 1991. Morphometric Tools for Landmark Data. Cambridge University Press.Google Scholar
  6. Bookstein, F.L. 1997. Landmark methods for forms without landmarks: Localizing group differences in outline shape. Medical Image Analysis, 1:225–243.Google Scholar
  7. Bregler,. 1997. Learning and recognizing human dynamics in video sequences. In Proc. Conf. Computer Vision and Pattern Recognition, pp. 568–574.Google Scholar
  8. Bulpitt, A.J. and Allinson, N.M. 1993. Human Motion Recognition Using Co-operative ART Networks. In Proc. World Congress on Neural Networks, Vol. 3, pp. 708–711.Google Scholar
  9. Buxton, H. and Gong, S. 1995. Visual suveillance in a dynamic and uncertain world. Artifical Intelligence, 78:431–459.Google Scholar
  10. Gumbel, E.J. 1958. Statistics of Extremes. Columbia University Press.Google Scholar
  11. Johnson, N. and Hogg, D.C. 1996. Learning the distribution of object trajectories for event recognition. Image and Vision Computing, 14:609–615.Google Scholar
  12. Monhaupt, M. and Neumann, B. 1990. On the use of motion concepts for top-down control in trafic scenes. In Proc. ECCV, pp. 598–600.Google Scholar
  13. Nagel, H.H. 1988. From image sequence towards conceptual descriptions. Image and Vision Computing, 6(2):59–74.Google Scholar
  14. Remagnino, P., Baumberg, A., Grove, T., Hogg, D., Tan, T., Worrall, A., and Baker, K. 1997. An intergrated traffic and pedestrian model-bassed vision system. In Proc. BMVC, pp. 380–389.Google Scholar
  15. Remagnino, P., Tan, T., and Baker, K. 1998. Agent orientated annotation in model based visual surveillance. In Proc. ICCV, pp. 857–862.Google Scholar
  16. Starner, T. and Pentland, A. 1995. Visual recognition of american sign language using hidden markov models. In Proc. Int. Workshop on Automatic Face and Gesture Recognition.Google Scholar
  17. Tan, T.N., Sullivan, K, G.D., and Baker, D. 1994. Recognising objects on the ground-plane. Image and Vision Computing, 12:164–172.Google Scholar

Copyright information

© Kluwer Academic Publishers 2000

Authors and Affiliations

  • R.J. Morris
    • 1
  • D.C. Hogg
    • 2
  1. 1.Department of StatisticsUniversity of LeedsEngland
  2. 2.School of Computer StudiesUniversity of LeedsEngland

Personalised recommendations