Interpretation of complex scenes using Bayesian networks

  • Mark F. Westling
  • Larry S. Davis
Poster Session II
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1352)


In most object recognition systems, interactions between objects in a scene are ignored and the best interpretation is considered to be the set of hypothesized objects that matches the greatest number of image features. Visual and physical interactions, however, provide a rich source of information: occlusion explains why features might be unde-tected, and physical constraints ensure a realisable interpretation. We show how these interations can be easily modeled using a Bayesian network, and how the problem of interpretation can be cast as finding the most likely explanation for such a network.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    R. Beveridge. A Local Search Algorithms for Geometric Object Recognition: Optimal Correspondence and Pose. PhD thesis, University of Massachussets, 1993.Google Scholar
  2. 2.
    T.O. Binford and T.S. Levitt. Model-based recognition of objects in complex scenes. In 1994 ARPA Image Understanding Workshop, November 1994.Google Scholar
  3. 3.
    T.O. Binford, T.S. Levitt, and W.B. Mann. Bayesian inference in model-based machine vision. Uncertainty in AI, 3:73–94, 1989.Google Scholar
  4. 4.
    T.M. Breuel. Higher-order statistics in object recognition. In CVPR93, pages 707–708. IEEE, 1993.Google Scholar
  5. 5.
    B. D'Ambrosio. SPI in large BN2O networks. In Proceedings of the Tenth Annual Conference on Uncertainty in Artificial Intelligence (UAI-94), pages 128–135, Seattle, WA, 1994.Google Scholar
  6. 6.
    C. Hansen and T. Henderson. CAGD-based computer vision. PAMI, 11(11):1181–1193, 1989.Google Scholar
  7. 7.
    D.P. Huttenlocher and S. Ullman. Recognizing solid objects by alignment with an image. IJCV, 5(2):195–212, November 1990.Google Scholar
  8. 8.
    T.S. Levitt, J.M. Agosta, and T.O. Binford. Model-based influence diagrams for machine vision. Uncertainty in AI, 5:233–244, 1990.Google Scholar
  9. 9.
    T.S. Levitt, T.O. Binford, G.J. Ettinger, and P. Gelband. Probability-based control for computer vision. In DARPA89, pages 355–369, 1989.Google Scholar
  10. 10.
    B. Li, Z. and D'Ambrosio. A framework for ordering composite beliefs. IEEE Transactions on Systems, Man, and Cybernetics, 25(2):243–255, 1995.Google Scholar
  11. 11.
    Z. Li and B. D'Ambrosio. Efficient inference in Bayes nets as a combinatorial optimization problem. Intl Journal of Approximate Reasoning, 11(1):55–81, 1994.Google Scholar
  12. 12.
    J. Liang, F. Jensen, and H. Christensen. A framework for generic object recognition with Bayesian networks. In Proceedings of the First International Symposium on Soft Computing for Pattern Recognition, Reading, U.K., March 1996.Google Scholar
  13. 13.
    W.B. Mann and T.O. Binford. SUCCESSOR: Interpretation overview and constraint system. In ARPA96, pages 1505–1518, 1996.Google Scholar
  14. 14.
    R. D. Rimey. Control of Selective Perception Using Bayes Nets and Decision Theory. PhD thesis, University of Rochester, December 1993.Google Scholar
  15. 15.
    C. Rothwell. Reasoning about occlusions during hypothesis verification. In ECCV96, volume 1, pages 599–609, 1996.Google Scholar
  16. 16.
    W. Wells. Statistical Pattern Recognition. PhD thesis, Massachussetts Institute of Technology, 1993.Google Scholar
  17. 17.
    M. Westling and L. Davis. Object recognition by fast hypothesis generation and reasoning about object interactions. In ICPR96, pages 148–153, Vienna, 1996.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1997

Authors and Affiliations

  • Mark F. Westling
    • 1
  • Larry S. Davis
    • 2
  1. 1.Perceptus TechnologiesBethesdaUSA
  2. 2.Computer Vision Laboratory, Center for Automation ResearchUniversity of MarylandCollege ParkUSA

Personalised recommendations