Skip to main content
SpringerLink
Log in

Discovering Shape Classes using Tree Edit-Distance and Pairwise Clustering

  • Published:
International Journal of Computer Vision Aims and scope Submit manuscript

Abstract

This paper describes work aimed at the unsupervised learning of shape-classes from shock trees. We commence by considering how to compute the edit distance between weighted trees. We show how to transform the tree edit distance problem into a series of maximum weight clique problems, and show how to use relaxation labeling to find an approximate solution. This allows us to compute a set of pairwise distances between graph-structures. We show how the edit distances can be used to compute a matrix of pairwise affinities using χ2 statistics. We present a maximum likelihood method for clustering the graphs by iteratively updating the elements of the affinity matrix. This involves interleaved steps for updating the affinity matrix using an eigendecomposition method and updating the cluster membership indicators. We illustrate the new tree clustering framework on shock-graphs extracted from the silhouettes of 2D shapes.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

References

  • Arcelli, C. and di Baja, G.S. 1992. Ridge points in euclidean distance maps. Pattern Recognition Letters, 13:237–243.

    Article  Google Scholar 

  • Barrow, H.G. and Burstall, R.M. 1976. Subgraph isomorphism, matching relational structures and maximal cliques. Inf. Proc. Letter, 4:83–84.

    Google Scholar 

  • Blum, H. 1973. Biological shape and visual science (Part I). Journal of theoretical Biology, 38:205–287.

    Article  MathSciNet  Google Scholar 

  • Bomze, I.M., Pelillo, M., and Stix, V. 2000. Approximating the maximum weight clique using replicator dynamics. IEEE Trans. on Neural Networks, vol. 11.

  • Bouix, S. and Siddiqi, K. 2000. Divergence-based medial surfaces. In Computer Vision ECCV 2000, Springer, LNCS 1842, pp. 603– 618.

  • Bridle, J. S. 1990. Training stochastic model recognition algorithms as networks can lead to maximum mutual information estimation of parameters. In Advances in Neural Information Processing Systems 2, San Mateo, CA: Morgan Kaufmann, pp. 211–217.

  • Bunke, H. and Kandel, A. 2000. Mean and maximum common subgraph of two graphs. Pattern Recognition Letters, 21:163–168.

    Article  Google Scholar 

  • Cox, T.F. and Cox, M.A.A. 1994. Multidimensional Scaling. Chapman & Hall.

  • Bowyer, K.W. and Dyer, C.R. 1990. Aspect graphs: An introduction and survey of recent results. International Journal of Imaging Systems and Technology, 2:315–328.

    Google Scholar 

  • Cyr, C.M. and Kimia, B.B. 2004. A similarity-based aspect-graph approach to 3D object recognition. International Journal of Computer Vision, 57(1):5–22.

    Article  Google Scholar 

  • Denton, T., Abrahamson, J., Shokoufandeh, A. 2004a. Approximation of canonical sets and their applications to 2D view simplification. In IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR’;04), vol. 2, pp. 550– 557.

  • Denton, T., Demirci, M., Abrahamson, J., Shokoufandeh, A., and Dickinson, S. 2004b. Selecting canonical views for view-based 3-D object recognition. In Proceedings, International Conference on Pattern Recognition (ICPR), Cambridge, U.K.

  • Dickinson, S., Pentland, A., and Rosenfeld, A. 1992. 3D shape recovery using distributed aspect matching. IEEE Trans. Pattern Anal. Machine Intell., 14(2):174–198.

    Google Scholar 

  • Eshera, M.A. and Fu, K.-S. 1986. An image understanding system using attributed symbolic representation and inexact graph-matching. IEEE Trans. Pattern Anal. Machine Intell., 8:604–618.

    Article  Google Scholar 

  • Ghahramani, Z. and Jordan, M.I. 1997. Factorial hidden markov models. Machine Learning, 29:245–273.

    Article  MATH  Google Scholar 

  • Gibbons, L.E. et al. 1997. Continuous characterizations of the maximum clique problem. Math. Oper. Res., 22:754–768.

    Article  MathSciNet  MATH  Google Scholar 

  • Giblin, P.J. and Kimia, B.B. 1999. On the local form and transitions of symmetry sets, medial axes, and shocks. In Proc. Int. Conf. Computer Vision, pp. 385–391.

  • Hofmann, T. and Buhmann, M. 1997. Pairwise data clustering by deterministic annealing. IEEE Trans. Pattern Anal. Machine Intell., 19(1):1–14.

    Article  Google Scholar 

  • Jain, A.K. and Dubes, R.C. 1988. Algorithms for Clustering Data. Prentice Hall: Englewood Cliffs, NJ.

    MATH  Google Scholar 

  • Kimia, B.B., Tannenbaum, A.R., and Zucker, S.W. 1995. Shapes, shocks, and deforamtions i. Int. J. Computer Vision, 15:189– 224.

    Article  Google Scholar 

  • Klein, P. et al. 1999. A tree-edit-distance algorithm for comparing simple, closed shapes. In ACM-SIAM Symp. on Disc.e Alg.

  • Lozano, M. and Escolano, F. 2003. EM Algorithm for Clustering and ensemble of graphs with comb matching. EMMVCPR, Lecture Notes in Computer Science, vol. 2683, pp. 52–67.

  • Luo, B., Wilson, R.C., and Hancock, E.R. 2003. Spectral embedding of graphs, Pattern Recognition, 36:2212–2223.

    Google Scholar 

  • Motzkin, T.S. and Straus, E.G. 1965. Maxima for graphs and a new proof of a theorem of Turán. Canadian Journal of Mathematics, 17:533–540.

    MathSciNet  MATH  Google Scholar 

  • Munger, A., Bunke, H., and Jiang, X. 1999. Combinatorial search vs. genetic algorithms: A case study based on the generalized median graph problem. Pattern Recognition Letters, 20(11–13):1271–1279.

    Google Scholar 

  • Ogniewicz, R.L. 1994. A multiscale mat from voronoi diagrams: the skeleton-space and its application to shape description and decomposition. In Aspects of Visual Form Processing, pp. 430–439.

  • Papadimitriou, C.H. and Steiglitz, K.J. 1982. Combinatorial optimization: Algorithms and Complexity. Prentice-Hall: Englewood Cliffs, NJ.

    MATH  Google Scholar 

  • Pavan, M. and Pelillo, M. 2003a. A new graph-theoretic approach to clustering and segmentation. In Proc. IEEE Conf. Computer Vision Pattern Recognition, vol. I, pp. 145–152.

  • Pavan, M. and Pelillo, M. 2003. Dominant sets and hierarchical clustering. In Proc. Int. Conf. Computer Vision, vol. I, pp. 362– 369.

  • Pelillo, M. 1997. The dynamics of relaxation labeling process, J. Math. Imaging Vision, 7:309–323.

    Google Scholar 

  • Pelillo, M. 1999. Replicator equations, maximal cliques, and graph isomorphism. Neural Computation, 11:1935–1955.

    Article  Google Scholar 

  • Pelillo, M., Siddiqi, K., and Zucker, S.W. 1999. Matching hierarchical structures using association graphs. IEEE Trans. Pattern Anal. Machine Intell., 21(11):1105–1120.

    Article  Google Scholar 

  • Pelillo, M. and Torsello, A. 2006. Payoff-monotonic game dynamics and the maximum clique problem. Neural Computation, 18:1215–1258.

    Article  MathSciNet  MATH  Google Scholar 

  • Perona, P. and Freeman, W.T. 1998. A factorization approach to grouping. In European Conference on Computer Vision, vol. 1, pp. 655–670.

  • Reyner, S.W. 1977. An analysis of a good algorithm for the subtree problem. SIAM Journal on Computing, 6:730–732.

    Article  MathSciNet  MATH  Google Scholar 

  • Rizzi, S. 1998. Genetic operators for hierarchical graph clustering. Pattern Recognition Letters, 19:1293–1300.

    Article  MATH  Google Scholar 

  • Sarkar, S. and Boyer, K.L. 1998. Quantitative measures of change based on feature organization: Eigenvalues and eigenvectors. Computer Vision and Image Understanding, 71(1):110–136.

    Article  Google Scholar 

  • Sebastian, T.S., Klein, P.N., and Kimia, B.B. 2001. Recognition of shapes by editing shock graphs. In Proc. Int. Conf. Computer Vision, vol. 1, pp. 755–762.

  • Sebastian, T.B., Klein, P.N., and Kimia, B.B. 2004. Recognition of shapes by editing their shock graphs. IEEE Trans. Pattern Anal. Machine Intell., 26(5):550–571.

    Article  Google Scholar 

  • Segen. J. 1988. Learning graph models of shape. In J. Laird (Ed.), Proceedings of the Fifth International Conference on Machine Learning, pp. 29–25.

  • Sengupta, K. and Boyer, K.L. 1995. Organizing large structural modelbases. IEEE Trans. Pattern Anal. Machine Intell., 17(4).

  • Sengupta, K. and Boyer, K.L. 1998. Modelbase partitioning using property matris spectra. Computer Vision and Image Understanding, 70(2).

  • Shi, J. and Malik, j. 2000. Normalized cuts and image segmentation. IEEE Trans. Pattern Anal. Machine Intell., 22(8):888–905.

    Article  Google Scholar 

  • Shokoufandeh, A. et al. 1999. Indexing using a spectral encoding of topological structure. In Proc. IEEE Conf. Computer Vision Pattern Recognition.

  • Siddiqi, K., Bouix, S., Tannenbaum, A., and Zucker, S.W. 1999a. The hamilton-jacobi skeleton. In Proc. Int. Conf. Computer Vision, pp. 828–834.

  • Siddiqi, K., Shokoufandeh, A., Dickinson, S.J., and Zucker, S.W. 1999b. Shock graphs and shape matching. Int. J. Computer Vision, 35(1):13–32.

    Article  Google Scholar 

  • Tirthapura, S. et al. 1998. Indexing based on edit-distance matching of shape graphs. In SPIE Int. Symp. on Voice, Video, and Data Comm., pp. 25–36.

  • Torsello, A. and Hancock, E.R. 2003. Computing approximate tree edit distance using relaxation labelling. Pattern Recognition Letters, 24:1089–1097.

    Article  MATH  Google Scholar 

  • Torsello, A., and Hancock, E.R. 2004. A skeletal measure of 2D shape similarity. Computer Vision and Image Understanding, 95(1):1–29.

    Article  Google Scholar 

  • Torsello, A. and Hancock, E.R. 2006. Learning shape-classes using a mixture of tree-unions. IEEE Trans. Pattern Anal. Machine Intell., 28(6):954–967.

    Article  Google Scholar 

  • Torsello, A., Hidovic-Rowe, D., and Pelillo, M. 2005. Polynomial-time metrics for attributed trees. IEEE Trans. Pattern Anal. Machine Intell., 27(7):1087– 1099.

    Article  Google Scholar 

  • Tsai, W.H. and Fu, K.S. 1979. Error-correcting isomorphism of attributed relational graphs for pattern analysis. IEEE Transactions on Systems, Man and Cybernetics, 9:757– 768.

    Article  MATH  Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Dipartimento di Informatica, University “Ca’ Foscari” of Venice, Via Torino 155, 30172, Venezia Mestre, Italy

    Andrea Torsello

  2. NICTA, Australian National University, RSISE Bldg. 115, Canberra, ACT, 0200, Australia

    Antonio Robles-Kelly

  3. Department of Computer Science, University of York, York, Y01 5DD, UK

    Edwin R. Hancock

Authors
  1. Andrea Torsello
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Antonio Robles-Kelly
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Edwin R. Hancock
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Andrea Torsello.

Additional information

National ICT Australia is funded by the Australian Government’s Backing Australia’s Ability initiative, in part through the Australian Research Council.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Torsello, A., Robles-Kelly, A. & Hancock, E.R. Discovering Shape Classes using Tree Edit-Distance and Pairwise Clustering. Int J Comput Vision 72, 259–285 (2007). https://doi.org/10.1007/s11263-006-8929-y

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11263-006-8929-y

Keywords

Search

Navigation