Skip to main content

An energy minimization method for matching and comparing structured object representations

  • Structural Models
  • Conference paper
  • First Online:
Energy Minimization Methods in Computer Vision and Pattern Recognition (EMMCVPR 1997)

Abstract

We present a general method for matching segmented parts of objects by energy minimization. The energy is designed in order to cope with possible imperfections of the compared segmentations (merged, or missing regions), and relies on the comparison of shape and positional descriptors. The minimization of the energy is performed by a simulated annealing procedure.

This work partially comes from a project on object recognition, supported by the D.R.E.T, helded at SUDIMAGE research laboratory. Publication is courtesy of the D.R.E.T.

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

Access this chapter

Institutional subscriptions

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. R. Azencott (1987): Image analysis and Markov random fields. Proc. of the int. Conf. on Ind. and Appl. Math. SIAM, Paris.

    Google Scholar 

  2. J. Ben-Arie and A. Z. Meiri (1987): 3D Object recognition by optimal search of multinary relation graphs Comp. Vis. Graph. Im. Proc. 37, 345–361

    Google Scholar 

  3. R. Bergerin and M. Levine (1993): Generic object recognition: building and matching coarse descriptions from line drawings IEEE Trans. Pat. Anal. Mach. Intel. Vol. 15, no 1, 19–36

    Google Scholar 

  4. O. Catoni (1990) Rough Large deviation estimates for simulated annealing. Application to exponential cooling schedules Ann. of Proba., 20, 1109–1146

    Google Scholar 

  5. W. E. L. Grimson and D. P. Huttenlocher (1991): On the verification of hypothetized matches in model-based recognition IEEE Trans. Pat. Anal. Mach. Intel.Vol 13, no 12, 1201–1213

    Google Scholar 

  6. A. R. Pope (1994): Model-based object recognition A survey of recent research Tech. Report 94-04

    Google Scholar 

  7. E. Rivlin, S. J. Dickinson, A. Rosenfeld (1994) Recognition by functional parts. (preprint Center for automation research).

    Google Scholar 

  8. L. G. Shapiro (1980) A structural model of shape IEEE Trans. Pat. Anal. Mach. Intel. vol 2 no 2 111–126

    Google Scholar 

  9. L. G. Shapiro and R. M. Haralik (1981): Structural description and inexact matching IEEE Trans. Pat. Anal. Mach. Intel. vol 3, no 5, 504–519

    Google Scholar 

  10. L. G. Shapiro and R. M. Haralik (1985): A metric for comparing relational descriptors IEEE Trans. Pat. Anal. Mach. Intel. vol 7 no 1 90–98

    Google Scholar 

  11. N. Ueda and S. Suzuki (1993): Learning visual models from shape contours using multi-scale convex/concave structure matching IEEE Trans. Pat. Anal. Mach. Intel. vol 15 no 4 337–351

    Google Scholar 

  12. B. Yang, W. E. Snyder and G. L. Bilbro (1989): Matching over-segmented 3D images to models using association graphs Image and vision comp. Vol 7 no 2 135–143

    Google Scholar 

  13. L. Younes (1996): Computable elastic distances between shapes (preprint).

    Google Scholar 

  14. S. Zhang, G. D. Sullivan, K. D. Baker (1993): The automatic construction of a view independent relational model for 3-D object recognition IEEE Trans. Pat. Anal. Mach. Intel. Vol 15 no 6 531–544

    Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Editor information

Marcello Pelillo Edwin R. Hancock

Rights and permissions

Reprints and permissions

Copyright information

© 1997 Springer-Verlag Berlin Heidelberg

About this paper

Cite this paper

Azencott, R., Younes, L. (1997). An energy minimization method for matching and comparing structured object representations. In: Pelillo, M., Hancock, E.R. (eds) Energy Minimization Methods in Computer Vision and Pattern Recognition. EMMCVPR 1997. Lecture Notes in Computer Science, vol 1223. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-62909-2_96

Download citation

  • DOI: https://doi.org/10.1007/3-540-62909-2_96

  • Published:

  • Publisher Name: Springer, Berlin, Heidelberg

  • Print ISBN: 978-3-540-62909-2

  • Online ISBN: 978-3-540-69042-9

  • eBook Packages: Springer Book Archive

Publish with us

Policies and ethics