Advertisement

Journal of Mathematical Imaging and Vision

, Volume 23, Issue 3, pp 345–365 | Cite as

On the Design of Filters for Gradient-Based Motion Estimation

  • Michael EladEmail author
  • Patrick Teo
  • Yacov Hel-Or
Article

Abstract

Gradient based approaches in motion estimation (Optical-Flow) refer to those techniques that estimate the motion of an image sequence based on local derivatives in the image intensity. In order to best evaluate local changes, specific filters are applied to the image sequence. These filters are typically composed of a spatiotemporal pre-smoothing filter followed by discrete derivative ones. The design of these filters plays an important role in the estimation accuracy. This paper proposes a method for such a design. Unlike previous methods that consider these filters as optimized approximations for continuum derivatives, the proposed design procedure defines the optimality directly with respect to the motion estimation goal. One possible result of the suggested scheme is a set of image dependent filters that can be computed prior to the estimation process. An alternative interpretation is the creation of generic filters, capable of treating natural images. Simulations demonstrate the validity of the new design approach.

Keywords

motion estimation optical flow pre-smoothing gradients computation optimal filters constrained minimization 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    R. Anderson and P. Bloomfield, “Numerical differentiation procedures for nonexact data,” Numerical Mathematics, Vol. 22, pp. 157–182, 1974.CrossRefGoogle Scholar
  2. 2.
    J.L. Barron, D.J. Fleet, and S. Beauchemin, “Performance of optical flow techniques,” International Journal of Comuter Vision, Vol. 12, pp. 43–77, 1994.CrossRefGoogle Scholar
  3. 3.
    T. Chin, W. Karl, and A. Willsky, “Probabilistic and sequential computation of optical flow using temporal coherence,” IEEE Trans. Image Processing, Vol. 3, pp. 773–788, 1994.CrossRefGoogle Scholar
  4. 4.
    M. Elad and A. Feuer, “Recursive optical flow estimation—adaptive filtering approach,” The Journal of Visual Comm. and Image Representation, Vol. 9, pp. 119–138, 1998.CrossRefGoogle Scholar
  5. 5.
    M. Elad, P. Teo, and Y. Hel-Or, “Optimal filters for gradient based motion estimation,” in International Conference on Computer Vision—ICCV 1999, Corfu, Greece, 1999, pp. 559–565.Google Scholar
  6. 6.
    D.J. Fleet and K. Langley, “Recursive filters for optical flow,” IEEE Trans. Pattern Analysis and Machine Inteligence (PAMI), Vol. 17, pp. 61–67, 1995.CrossRefGoogle Scholar
  7. 7.
    R.A. Horn and C.J. Johnson, Matrix Analysis, 1st ed., Cambridge University Press, 1985.Google Scholar
  8. 8.
    B.K.P. Horn and B.G. Schunck, “Determining optical flow,” Artificial Inteligence, Vol. 17, pp. 185–203, 1981.CrossRefGoogle Scholar
  9. 9.
    B. Jahne, Digital Image Processing—Concepts, Algorithms, and Scientific Applications, 3rd ed., Springer-Verlag, 1995.Google Scholar
  10. 10.
    B. Lucas and T. Kanade, “An iterative image registration technique with an application to stereo vision,” in Proc. DARPA, Image Understanding Workshop, 1981, pp. 121–13.Google Scholar
  11. 11.
    B. Porat, A Course in Digital Signal Processing, 1st edn., Wiley, 1997.Google Scholar
  12. 12.
    D. Robinson and P. Milanfar, “Fundamental performance limits in image registration,” IEEE Transactions on Image Processing, Vol. 13, No. 9, pp. 1185–1199, 2004.CrossRefPubMedGoogle Scholar
  13. 13.
    D. Robinson and P. Milanfar, “Bias minimizing filters for gradient-based motion estimation,” in Proceedings of the 37th Asilomar Conference on Signals, Systems and Computers, Pacific Grove, CA, November, 2004.Google Scholar
  14. 14.
    M. Shin, D. Goldgof, K. Bowyer, and S. Nikiforou, “Comparison of edge detection algorithms using a structure from motion task,” IEEE Trans. Systems, Man, and Cybernetics—Part B, Vol. 31, No. 4, 2001.Google Scholar
  15. 15.
    E.P. Simoncelli, “Design of multi-dimensional derivatives filters,” in The IEEE International Conf. on Image Processing, Austin Tx, 1994.Google Scholar
  16. 16.
    A. Singh, Optic Flow Computation: A unified Perspective, 1st edn., IEEE Computer Society Press: California, 1991.Google Scholar
  17. 17.
    E. Simoncelli, W. Freeman, E. Adelson, and D. Heeger, “Shiftable multiscale transforms,” IEEE Trans. Information Theory, Vol. 38, pp. 587–607, 1992.CrossRefGoogle Scholar

Copyright information

© Springer Science + Business Media, Inc. 2005

Authors and Affiliations

  1. 1.Department of Computer ScienceThe TechnionHaifaIsrael
  2. 2.Redwood City
  3. 3.Department of Computer ScienceInter-Disciplinary CenterHerzliyaIsrael

Personalised recommendations