Aligning Shapes by Minimising the Description Length

  • Anders Ericsson
  • Johan Karlsson
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3540)

Abstract

When building shape models, it is first necessary to filter out the similarity transformations from the original configurations. This is normally done using Procrustes analysis, that is minimising the sum of squared distances between the corresponding landmarks under similarity transformations. In this article we propose to align shapes using the Minimum Description Length (MDL) criterion. Previously MDL has been used to locate correspondences. We show that the Procrustes alignment with respect to rotation is not optimal.

The MDL based algorithm is compared with Procrustes on a number of data sets. It is concluded that there is improvement in generalisation when using Minimum Description Length. With a synthetic example it is shown that the Procrustes alignment can fail significantly where the proposed method does not.

The Description Length is minimised using Gauss-Newton. In order to do this the derivative of the description length with respect to rotation is derived.

Keywords

Similarity Transformation Shape Model Minimum Description Length Goal Function Latin Letter 
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.
    Andrew, A., Chu, E., Lancaster, P.: Derivatives of eigenvalues and eigenvectors of matrix functions, pp. 903–926 (1993)Google Scholar
  2. 2.
    Bookstein, F.L.: Size and shape spaces for landmark data in two dimensions. Statistical Science 1(2), 181–242 (1986)MATHCrossRefGoogle Scholar
  3. 3.
    Davies, R.H., Twining, C.J., Cootes, T.F., Waterton, J.C., Taylor, C.J.: A minimum description length approach to statistical shape modeling. IEEE Trans. medical imaging 21(5), 525–537 (2002)CrossRefGoogle Scholar
  4. 4.
    Davies, R.H., Cootes, T.F., Waterton, J.C., Taylor, C.J.: An efficient method for constructing optimal statistical shape models. In: Niessen, W.J., Viergever, M.A. (eds.) MICCAI 2001. LNCS, vol. 2208, pp. 57–65. Springer, Heidelberg (2001)Google Scholar
  5. 5.
    Dryden, I.L., Mardia, K.V.: Statistical Shape Analysis. John Wiley & Sons, Inc., Chichester (1998)MATHGoogle Scholar
  6. 6.
    Ericsson, A.: Automatic Shape Modelling and Applications in Medical Imaging. PhD thesis, Mathematics LTH, Lund University, Centre for Mathematical Sciences, Box 118, SE-22100, Lund, Sweden (November 2003)Google Scholar
  7. 7.
    Ericsson, A., Åström, K.: Minimizing the description length using steepest descent. In: Proc. British Machine Vision Conference, Norwich, United Kingdom, vol. 2, pp. 93–102 (2003)Google Scholar
  8. 8.
    Gower, J.C.: Generalized procrustes analysis. Psychometrica 40, 33–50 (1975)MATHCrossRefMathSciNetGoogle Scholar
  9. 9.
    Karlsson, J., Ericsson, A., Åström, K.: Parameterisation invariant statistical shape models. In: Proc. International Conference on Pattern Recognition, Cambridge, UK (2004)Google Scholar
  10. 10.
    Kendall, D.G., Barden, D., Carne, T.K., Le, H.: Shape and Shape Theory. John Wiley & Sons Ltd., Chichester (1999)MATHCrossRefGoogle Scholar
  11. 11.
    Papadopoulo, T., Lourakis, M.: Estimating the jacobian of the singular value decomposition. In: Vernon, D. (ed.) ECCV 2000. LNCS, vol. 1842, pp. 555–559. Springer, Heidelberg (2000)CrossRefGoogle Scholar
  12. 12.
    Rissanen, J.: Modeling by shortest data description. Automatica 14, 465–471 (1978)MATHCrossRefGoogle Scholar
  13. 13.
    Small, C.G.: The Statistical Theory of Shape. Springer, Heidelberg (1996)MATHGoogle Scholar
  14. 14.
    Thodberg, H.H.: Minimum description length shape and appearance models. In: Taylor, C.J., Noble, J.A. (eds.) IPMI 2003. LNCS, vol. 2732, pp. 51–62. Springer, Heidelberg (2003)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2005

Authors and Affiliations

  • Anders Ericsson
    • 1
  • Johan Karlsson
    • 1
  1. 1.Centre for Mathematical SciencesLund UniversityLundSweden

Personalised recommendations