Advertisement

Quantitative Evaluation on Heat Kernel Permutation Invariants

  • Bai Xiao
  • Richard C. Wilson
  • Edwin R. Hancock
Part of the Lecture Notes in Computer Science book series (LNCS, volume 5342)

Abstract

The Laplacian spectrum has proved useful for pattern analysis tasks, and one of its important properties is its close relationship with the heat equation. In this paper, we first show how permutation invariants computed from the trace of the heat kernel can be used to characterize graphs for the purposes of measuring similarity and clustering. We explore three different approaches to characterize the heat kernel trace as a function of time. These are the heat kernel trace moments, heat content invariants and symmetric polynomials with Laplacian eigenvalues as inputs. We then use synthetic and real world datasets to give a quantitative evaluation of these feature invariants deduced from heat kernel analysis. We compare their performance with traditional spectrum invariants.

Keywords

Zeta Function Heat Kernel Random Graph Symmetric Polynomial Edit Operation 
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.
    Atkins, J.E., Bowman, E.G., Hendrickson, B.: A spectral algorithm for seriation and the consecutive ones problems. SIAM J. Comput. 28, 297–310 (1998)MathSciNetCrossRefzbMATHGoogle Scholar
  2. 2.
    Chung, F.R.K.: Spectral graph theory. American Mathematical Society (1997)Google Scholar
  3. 3.
    Figueiredo, M.A.F., Jain, A.K.: Unsupervised learning of finite mixture models. IEEE Trans. PAMI 24(3), 381–396 (2002)CrossRefGoogle Scholar
  4. 4.
    Chaval, I., Farkas, H.M.: Extremal kahler metrics ii. Differential Geometry and Complex Analysis, 95–114 (1985)Google Scholar
  5. 5.
    Shi, J., Malik, J.: Normalized cuts and image segmentation. IEEE Trans. PAMI 22, 888–905 (2000)CrossRefGoogle Scholar
  6. 6.
    McDonald, P., Meyers, R.: Diffusions on graphs, poisson problems and spectral geometry. Transactions on American Mathematical Society 354, 5111–5136 (2002)MathSciNetCrossRefzbMATHGoogle Scholar
  7. 7.
    Mori, G., Belongie, S.J., Malik, J.: Efficient shape matching using shape contexts. IEEE Trans. Pattern Anal. Mach. Intell. 27(11), 1832–1837 (2005)CrossRefzbMATHGoogle Scholar
  8. 8.
    Sebastian, T.B., Klein, P.N., Kimia, B.B.: Recognition of shapes by editing their shock graphs. IEEE Trans. Pattern Anal. Mach. Intell. 26(5), 550–571 (2004)CrossRefGoogle Scholar
  9. 9.
    Torsello, A., Robles-Kelly, A., Hancock, E.R.: Discovering shape classes using tree edit-distance and pairwise clustering. International Journal on Computer Vision 72(3), 259–285 (2007)CrossRefGoogle Scholar
  10. 10.
    Wilson, R.C., Luo, B., Hancock, E.R.: Pattern vectors from algebraic graph theory. IEEE Trans. PAMI 27, 2220–2237 (2003)Google Scholar
  11. 11.
    Xiao, B., Hancock, E.R.: Clustering shapes using heat content invariants. In: ICIP (1), pp. 1169–1172 (2005)Google Scholar
  12. 12.
    Xiao, B., Hancock, E.R.: Trace formula analysis of graphs. In: SSPR/SPR, pp. 306–313 (2006)Google Scholar
  13. 13.
    Yau, S.T., Schoen, R.M.: Differential geometry. Science Publication (1988)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2008

Authors and Affiliations

  • Bai Xiao
    • 1
  • Richard C. Wilson
    • 2
  • Edwin R. Hancock
    • 2
  1. 1.Department of Computer ScienceUniversity of BathBathUK
  2. 2.Department of Computer ScienceUniversity of YorkYorkUK

Personalised recommendations