A theory of occlusion in the context of optical flow

  • Steven S. Beauchemin
  • John L. Barron
Part of the Advances in Computing Science book series (ACS)


Traditionally, image motion and its approximation known as optical flow have been treated as continuous functions of the image domain [9]. However, in realistic imagery, one finds cases verifying this hypothesis exceedingly rarely. Many phenomena may cause discontinuities in the optical flow function of imagery [16]. Among them, occlusion and translucency are frequent causes of discontinuities in realistic imagery. In addition, their information content is useful to later stages of processing [8] such as motion segmentation [1] and 3-d surface reconstruction [17].


Optical Flow Depth Function Intensity Profile Fourier Spectrum Visual Sensor 
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]
    G. Adiv. Determining three-dimensional motion and structure from optical flow generated by several moving objects. IEEE PAMI, 7 (4): 384–401, 1985.CrossRefGoogle Scholar
  2. [2]
    S. S. Beauchemin and J. L. Barron. The computation of optical flow. ACM Computing Surveys, 27 (3): 433–467, 1995.CrossRefGoogle Scholar
  3. [3]
    S. S. Beauchemin and J. L. Barron. A theory of occlusion. Technical Report TR-449, Dept. of Computer Science, Univ. of Western Ontario, March 1995.Google Scholar
  4. [4]
    S. S. Beauchemin, A. Chalifour, and J. L. Barron. Discontinuous optical flow: Recent theoretical results. In Vision Interface, pages 57–64, Kelowna, Canada, May 1997.Google Scholar
  5. [5]
    J. R. Bergen, P. J. Burt, R. Hingorani, and S. Peleg. Three-frame algorithm for estimating two-component image motion. IEEE PAMI, 14 (9): 886–896, 1992.CrossRefGoogle Scholar
  6. [6]
    M. J. Black and P. Anandan. A model for the detection of motion over time. In Proceedings of ICCV, pages 33–37, Osaka, Japan, December 1990.Google Scholar
  7. [7]
    T. Darrell and A. Pentland. Robust estimation of a multi-layered motion representation. In IEEE Proceedings of Workshop on Visual Motion, pages 173–178, Princeton, New Jersey, October 1991.CrossRefGoogle Scholar
  8. [8]
    D. J. Fleet. Measurement of Image Velocity. Kluwer Academic Publishers, Norwell, 1992.MATHCrossRefGoogle Scholar
  9. [9]
    B. K. P. Horn and B. G. Schunck. Determining optical flow. Artificial Intelligence, 17: 185–204, 1981.CrossRefGoogle Scholar
  10. [10]
    M. Irani, B. Rousso, and S. Peleg. Computing occluding and transparent motions. IJCV, 12 (1): 5–16, 1994.CrossRefGoogle Scholar
  11. [11]
    A. D. Jepson and M. Black. Mixture models for optical flow computation. In IEEE Proceedings of CVPR, pages 760–761, New York, New York, June 1993.Google Scholar
  12. [12]
    D. Marr and S. Ullman. Directional selectivity and its use in early visual processing. Proceedings of Royal Society London, B 211: 151–180, 1981.CrossRefGoogle Scholar
  13. [13]
    H.-H. Nagel. Displacement vectors derived from second-order intensity variations in image sequences. CVGIP, 21: 85–117, 1983.Google Scholar
  14. [14]
    C. Schnorr. Computation of discontinuous optical flow by domain decomposition. IEEE PAMI, 8 (2): 153–165, 1992.Google Scholar
  15. [15]
    B. G. Schunck. Image flow segmentation and estimation by constraint line clustering. IEEE PAMI, 11 (10): 1010–1027, 1989.CrossRefGoogle Scholar
  16. [16]
    M. Shizawa and K. Masc. Principle of superposition: A common computational framework for analysis of multiple motion. In IEEE Proceedings of Workshop on Visual Motion, pages 164–172, Princeton, New Jersey, October 1991.CrossRefGoogle Scholar
  17. [17]
    P. Toh and A. K. Forrest. Occlusion detection in early vision. In ICCV, pages 126–132. IEEE, 1990.Google Scholar

Copyright information

© Springer-Verlag/Wien 1997

Authors and Affiliations

  • Steven S. Beauchemin
  • John L. Barron

There are no affiliations available

Personalised recommendations