Optical Flow Diffusion with Robustified Kernels

  • Ashish Doshi
  • Adrian G. Bors
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3691)


This paper provides a comparison study among a set of robust diffusion algorithms for processing optical flows. The proposed algorithms combine the smoothing ability of the heat kernel, modelled by the local Hessian, and the outlier rejection mechanisms of robust statistics algorithms. Smooth optical flow variation can be modelled very well using heat kernels. The diffusion kernel is considered Gaussian, where the covariance matrix implements the inverse of the local Hessian. Robust statistics operators improve the results provided by the heat kernel based diffusion, by rejecting outliers and by avoiding optical flow oversmoothing. Alpha-trimmed mean and median statistics are considered for robustifying diffusion kernels. The robust diffusion smoothing is applied onto multiple frames and is extended to 3D lattices.


Mean Square Error Optical Flow Motion Vector Heat Kernel Radial Basis Function Network 
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.
    Tekalp, A.M.: Digital Video Processing. Prentice Hall, Englewood Cliffs (1995)Google Scholar
  2. 2.
    Barron, J.L., Fleet, D.J., Beauchemin, S.: Performance of optical flow techniques. International Journal of Computer Vision 12(1), 43–77 (1994)CrossRefGoogle Scholar
  3. 3.
    Black, M., Anandan, P.: The robust estimation of multiple motions: parametric and piecewise-smooth flow fields. Computer Vision and Image Understanding 63(1), 75–104 (1996)CrossRefGoogle Scholar
  4. 4.
    Bors, A.G., Pitas, I.: Optical Flow Estimation and Moving Object Segmentation Based on Median Radial Basis Function Network. IEEE Trans. on Image Processing 7(5), 693–702 (1998)CrossRefGoogle Scholar
  5. 5.
    Perona, P., Malik, J.: Scale-Space and Edge Detection Using Anisotropic Diffusion. IEEE Trans. on Pattern Anal. and Machine Intell. 12(7), 629–639 (1990)CrossRefGoogle Scholar
  6. 6.
    Yau, S.-T. (ed.): Surveys in Differential Geometry: Differential Geometry inspired by String Theory. American Mathematical Society (1999)Google Scholar
  7. 7.
    Black, M.J., Sapiro, G., Marimont, D.H., Heeger, D.: Robust Anisotropic Diffusion. IEEE Trans. on Image Processing 7(3), 421–432 (1998)CrossRefGoogle Scholar
  8. 8.
    Tschumperlé, D., Deriche, R.: Orthonormal vector sets regularization with PDE’s and Applications. Inter. Jour. on Computer Vision 50(3), 237–252 (2002)zbMATHCrossRefGoogle Scholar
  9. 9.
    Tschumperlé, D., Deriche, R.: Vector-Valued Image Regularization with PDE’s: A Common Framework for Different Applications. In: Proc. IEEE Computer Vision and Pattern Recognition, Madison, USA, vol. I, pp. 651–656 (2003)Google Scholar
  10. 10.
    Burgi, P.-Y.: Motion estimation based on the direction of intensity gradient. Image and Vision Computing 22(8), 637–653 (2004)CrossRefGoogle Scholar
  11. 11.
    Bertalmio, M., Vese, L., Sapiro, G., Osher, S.: Simultaneous Structure and Texture Image Inpainting. IEEE Trans. on Image Proces. 12(8), 882–889 (2003)CrossRefGoogle Scholar
  12. 12.
    Irani, M.: Multi-Frame Optical Flow Estimation Using Subspace Constraints. In: Proc. IEEE Int. Conf. on Computer Vision, Corfu, Greece, vol. I, pp. 626–633 (1999)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2005

Authors and Affiliations

  • Ashish Doshi
    • 1
  • Adrian G. Bors
    • 1
  1. 1.Dept. of Computer ScienceUniversity of YorkYorkUnited Kingdom

Personalised recommendations