Discovering Shape Categories by Clustering Shock Trees

  • B. Luo
  • A. Robles-Kelly
  • A. Torsello
  • R. C. Wilson
  • E. R. Hancock
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 2124)


This paper investigates whether meaningful shape categories can be identified in an unsupervised way by clustering shock-trees. We commence by computing weighted and unweighted edit distances between shock-trees extracted from the Hamilton-Jacobi skeleton of 2D binary shapes. Next we use an EM-like algorithm to locate pairwise clusters in the pattern of edit-distances. We show that when the tree edit distance is weighted using the geometry of the skeleton, then the clustering method returns meaningful shape categories.


clustering shock trees EM algorithm Hamilton-Jacobi skeleton 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Siddiqi, K., Bouix, S., Tannenbaum, A., Zucker, S.W.: The hamilton-jacobi skeleton. ICCV (1999) 828–834Google Scholar
  2. 2.
    Siddiqi, K., Shokoufandeh, A., Dickinson, S.J., Zucker, S.W.: Shock graphs and shape matching. International Journal of Computer Vision, 35 (1999) 13–32CrossRefGoogle Scholar
  3. 3.
    Kimia, B.B., Tannenbaum, A.R., Zucker, S.W.: Shapes, shocks, and deforamtions i. International Journal of Computer Vision, 15 (1995) 189–224CrossRefGoogle Scholar
  4. 4.
    Rizzi, S.: Genetic operators for hierarchical graph clustering. Pattern Recognition Letters, 19 (1998) 1293–1300zbMATHCrossRefGoogle Scholar
  5. 5.
    Segen, J.: Learning graph models of shape. In: Laird, J. (ed.): Proceedings of the Fifth International Conference on Machine Learning (1988) 29–25Google Scholar
  6. 6.
    Sengupta, K., Boyer, K.L.: Organizing large structural modelbases. IEEE Transactions on Pattern Analysis and Machine Intelligence, 17 (1995)Google Scholar
  7. 7.
    Munger, A., Bunke, H., Jiang, X.: Combinatorial search vs. genetic algorithms: A case study based on the generalized median graph problem. Pattern Recognition Letters, 20 (1999) 1271–1279CrossRefGoogle Scholar
  8. 8.
    Shapiro, L.G., Haralick, R.M.: Relational models for scene analysis. IEEE Transactions on Pattern Analysis and Machine Intelligence, 4 (1982) 595–602zbMATHCrossRefGoogle Scholar
  9. 9.
    Eshera, M.A., Fu, K.S.: A graph distance measure for image analysis. IEEE Transactions on Systems, Man and Cybernetics, 14 (1984) 398–407zbMATHGoogle Scholar
  10. 10.
    Kittler, J., Christmas, W.J., Petrou, M.: Structural matching in computer vision using probabilistic relaxation. IEEE PAMI, 17 (1995) 749–764Google Scholar
  11. 11.
    Wilson, R., Hancock, E.R.: Structural matching by discrete relaxation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 19 (1997) 634–648CrossRefGoogle Scholar
  12. 12.
    Huet, B., Hancock, E.: Relational histograms for shape indexing. In: IEEE International Conference of Computer Vision (1998)Google Scholar
  13. 13.
    Sengupta, K., Boyer, K.L.: Modelbase partitioning using property matris spectra. Computer Vision and Image Understanding, 70 (1998)Google Scholar
  14. 14.
    Hofmann, T., Buhmann, M.: Pairwise data clustering by deterministic annealing. IEEE Tansactions on Pattern Analysis and Machine Intelligence, 19 (1997)Google Scholar
  15. 15.
    Shokoufandeh, A., et al.: Indexing using a spectral encoding of topological structure. In: Conference on Computer Vision and Pattern Recognition (June 1999)Google Scholar
  16. 16.
    Torsello, A., Hancock, E.R.: A skeletal measure of 2d shape similarity. In: Visual Form 2001. LNCS 2059 (2001)CrossRefGoogle Scholar
  17. 17.
    Sarkar, S., Boyer, K.L.: Quantitative measures of change based on feature organization: Eigenvalues and eigenvectors. Computer Vision and Image Understanding, 71 (1998) 110–136CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2001

Authors and Affiliations

  • B. Luo
    • 1
  • A. Robles-Kelly
    • 1
  • A. Torsello
    • 1
  • R. C. Wilson
    • 1
  • E. R. Hancock
    • 1
  1. 1.Department of Computer ScienceUniversity of YorkYorkUK

Personalised recommendations