Segmentation and estimation of the optical flow

  • Adrian G. Borş
  • Ioannis Pitas
Posters
Part of the Lecture Notes in Computer Science book series (LNCS, volume 970)

Abstract

We propose an algorithm for simultaneous estimation and segmentation of the optical flow. The moving scene is decomposed in different regions with respect to their motion, by means of a pattern recognition scheme. The feature vectors are drawn from the image sequence and they are used to train a Radial Basis Functions (RBF) neural network. The learning algorithm for the RBF network minimizes a cost function derived from the probability estimation theory. The proposed algorithm was applied in real image sequences.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    A. N. Netravali, B. G. Haskell, Digital Pictures, Representation and compression, (Applications of Communications Theory, series editor: R. W. Lucky), Plenum Press, 1988.Google Scholar
  2. 2.
    J. Konrad, E. Dubois, “Bayesian estimation of motion vector fields” IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. PAMI-14, No. 9, pp. 910–927, Sep. 1992.Google Scholar
  3. 3.
    J. Besag, “On the statistical analysis of dirty pictures,” J. Roy Statist. Soc., vol. 48, ser. B, No. 3, pp. 259–302, 1986.Google Scholar
  4. 4.
    F. Heitz, P. Bouthemy, “Multimodal estimation of discontinuous optical flow using Markov Random Fields,” IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. PAMI-15, No. 12, Dec. 1993.Google Scholar
  5. 5.
    M. M. Chan, M. I. Sezan, A. M. Tekalp, “An algorithm for simultaneous motion estimation and scene segmentation,” Proc. IEEE Int. Conf. Acoust. Speech and Signal Processing, Adelaide, Australia, pp. V-221–V-224, May 1994.Google Scholar
  6. 6.
    A. G. Borş, M. Gabbouj, “Minimal topology for a radial basis functions neural network for pattern classification,” Digital Signal Processing, vol. 4, no. 3, pp. 173–188, July 1994.Google Scholar
  7. 7.
    A. G. Borş, I. Pitas, “Robust estimation of radial basis functions,” Proc. of IEEE Workshop on Neural Networks for Signal Processing IV, Ermioni, Greece, pp. 105–114, 6–8 Sep. 1994.Google Scholar
  8. 8.
    R. J. Schalkoff, Pattern Recognition: Statistical, Structural and Neural Approaches, John Wiley, 1992.Google Scholar
  9. 9.
    A. G. Borş, I. Pitas, “Median radial basis functions network for optical flow processing,” Proc. of IEEE Workshop on Nonlinear Signal and Image Processing, Neos Marmaras, Greece, pp. 702–705, 20–22 June 1995.Google Scholar

Copyright information

© Springer-Verlag 1995

Authors and Affiliations

  • Adrian G. Borş
    • 1
  • Ioannis Pitas
    • 1
  1. 1.Department of InformaticsUniversity of ThessalonikiThessalonikiGreece

Personalised recommendations