Advertisement

Abstract

Shape matching plays a prominent role in the analysis of medical and biological structures. Recently, a unifying framework was introduced for shape matching that uses mixture-models to couple both the shape representation and deformation. Essentially, shape distances were defined as geodesics induced by the Fisher-Rao metric on the manifold of mixture-model represented shapes. A fundamental drawback of the Fisher-Rao metric is that it is NOT available in closed-form for the mixture model. Consequently, shape comparisons are computationally very expensive. Here, we propose a new Riemannian metric based on generalized φ- entropy measures. In sharp contrast to the Fisher-Rao metric, our new metric is available in closed-form. Geodesic computations using the new metric are considerably more efficient. Discriminative capabilities of this new metric are studied by pairwise matching of corpus callosum shapes. Comparisons are conducted with the Fisher-Rao metric and the thin-plate spline bending energy.

Keywords

Gaussian Mixture Model Fisher Information Matrix Christoffel Symbol Shape Match Shape Representation 
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.

References

  1. 1.
    Small, C.: The statistical theory of shape. Springer, New York (1996)MATHGoogle Scholar
  2. 2.
    Bookstein, F.L.: Principal warps: Thin-plate splines and the decomposition of deformations. IEEE Trans. Patt. Anal. Mach. Intell. 11, 567–585 (1989)MATHCrossRefGoogle Scholar
  3. 3.
    Rohr, K., Stiehl, H., Sprengel, R., Buzug, T., Weese, J., Kuhn, M.: Landmark-based elastic registration using approximating thin-plate splines. IEEE Trans. on Medical Imaging 20, 526–534 (2001)CrossRefGoogle Scholar
  4. 4.
    Davies, R., Twining, C., Cootes, T., Taylor, C.: An information theoretic approach to statistical shape modelling. In: Heyden, A., Sparr, G., Nielsen, M., Johansen, P. (eds.) ECCV 2002. LNCS, vol. 2351, pp. 3–20. Springer, Heidelberg (2002)Google Scholar
  5. 5.
    Camion, V., Younes, L.: Geodesic interpolating splines. In: Figueiredo, M., Zerubia, J., Jain, A.K. (eds.) EMMCVPR 2001. LNCS, vol. 2134, pp. 513–527. Springer, Heidelberg (2001)CrossRefGoogle Scholar
  6. 6.
    Joshi, S., Miller, M.: Landmark matching via large deformation diffeomorphisms. IEEE Trans. Image Processing 9, 1357–1370 (2000)MATHCrossRefMathSciNetGoogle Scholar
  7. 7.
    Chui, H., Rangarajan, A.: A new point matching algorithm for non-rigid registration. Computer Vision and Image Understanding 89, 114–141 (2003)MATHCrossRefGoogle Scholar
  8. 8.
    Guo, H., Rangarajan, A., Joshi, S., Younes, L.: Non-rigid registration of shapes via diffeomorphic point matching. In: ISBI 2004 (2004)Google Scholar
  9. 9.
    Siddiqi, K., Shokoufandeh, A., Dickinson, S.J., Zucker, S.W.: Shock graphs and shape matching. In: ICCV, pp. 222–229 (1998)Google Scholar
  10. 10.
    Peter, A., Rangarajan, A.: Shape matching using the Fisher-Rao Riemannian metric: Unifying shape representation and deformation. In: ISBI 2006 (2006)Google Scholar
  11. 11.
    Burbea, J., Rao, R.: Entropy differential metric, distance and divergence measures in probability spaces: A unified approach. Journal of Multivariate Analysis 12, 575–596 (1982)MATHCrossRefMathSciNetGoogle Scholar
  12. 12.
    Havrda, M.E., Charvát, F.: Quantification method of classification processes: Concept of structural α-entropy. Kybernetica 3, 30–35 (1967)MATHGoogle Scholar
  13. 13.
    Rao, C.: Information and accuracy attainable in estimation of statistical parameters. Bulletin of the Calcutta Mathematical Society 37, 81–91 (1945)MATHMathSciNetGoogle Scholar
  14. 14.
    Costa, S., Santos, S., Strapasson, J.: Fisher information matrix and hyperbolic geometry. In: IEEE Information Theory Workshop, pp. 28–30 (2005)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Adrian Peter
    • 1
  • Anand Rangarajan
    • 2
  1. 1.Dept. of ECEUniversity of FloridaGainesville
  2. 2.Dept. of CISEUniversity of FloridaGainesville

Personalised recommendations