A comparison of stochastic and deterministic solution methods in Bayesian estimation of 2-D motion

  • Janusz Konrad
  • Eric Dubois
Optical Flow
Part of the Lecture Notes in Computer Science book series (LNCS, volume 427)


A new stochastic motion estimation method based on the Maximum A Posteriori Probability (MAP) criterion is developed. Deterministic algorithms approximating the MAP estimation over discrete and continuous state spaces are proposed. These approximations result in known motion estimation algorithms. The theoretical superiority of the stochastic algorithms over deterministic approximations in locating the global optimum is confirmed experimentally.


Mean Square Error Gibbs Sampler Motion Field Motion Estimation Algorithm Deterministic Approximation 
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.


  1. [1]
    D.W. Murray and B.F. Buxton, “Scene segmentation from visual motion using global optimization,” IEEE Trans. Pattern Anal. and Mach. Intell., vol. PAMI-9, pp. 220–228, March 1987.Google Scholar
  2. [2]
    J. Konrad and E. Dubois, “Estimation of image motion fields: Bayesian formulation and stochastic solution,” in Proc. IEEE Int. Conf. on Acoust., Speech, and Signal Process. ICASSP'88, 1988, pp. 1072–1074.Google Scholar
  3. [3]
    J. Konrad and E. Dubois, “Multigrid Bayesian estimation of image motion fields using stochastic relaxation,” in Proc. IEEE Int. Conf. Computer Vision ICCV'88, 1988, pp. 354–362.Google Scholar
  4. [4]
    J. Konrad and E. Dubois, “Bayesian estimation of discontinuous motion in images using simulated annealing,” in Proc. Conf. Vision Interface VI'89, 1989, pp. 51–60.Google Scholar
  5. [5]
    P. Bouthemy and P. Lalande, “Motion detection in an image sequence using Gibbs distributions,” in Proc. IEEE Int. Conf. on Acoust., Speech, and Signal Process. ICASSP'89, 1989, pp. 1651–1654.Google Scholar
  6. [6]
    F. Heitz and P. Bouthemy, “Estimation et segmentation du mouvement: approache bayésienne et modelisation markovienne des occlusions,” in Proc. of 7-th Congress AFCET — RFIA, 1989.Google Scholar
  7. [7]
    J. Hutchinson, Ch. Koch, J. Luo and C. Mead, “Computing motion using analog and binary resistive networks,” Computer, vol. 21, pp. 52–63, March 1988.Google Scholar
  8. [8]
    S. Kirkpatrick, C.D. Gelatt, Jr. and M.P. Vecchi, “Optimization by simulated annealing,” Science, vol. 220, pp. 671–680, May 1983.Google Scholar
  9. [9]
    S. Geman and D. Geman, “Stochastic relaxation, Gibbs distributions, and the Bayesian restoration of images,” IEEE Trans. Pattern Anal. and Mach. Intell., vol. PAMI-6, pp. 721–741, November 1984.Google Scholar
  10. [10]
    J. Konrad, “Bayesian estimation of motion fields from image sequences,” Ph.D. Thesis, McGill University, Dept. of Electr. Eng., 1989.Google Scholar
  11. [11]
    B.K.P. Horn and B.G. Schunck, “Determining optical flow,” Artificial Intelligence, vol. 17, pp. 185–203, 1981.Google Scholar
  12. [12]
    J. Besag, “On the statistical analysis of dirty pictures,” J. R. Statist. Soc., vol. 48, B, pp. 259–279, 1986.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1990

Authors and Affiliations

  • Janusz Konrad
    • 1
  • Eric Dubois
    • 1
  1. 1.INRS-TélécommunicationsVerdunCanada

Personalised recommendations