Skip to main content

A Comparison of Nearest Neighbor Search Algorithms for Generic Object Recognition

  • Conference paper
Advanced Concepts for Intelligent Vision Systems (ACIVS 2006)

Part of the book series: Lecture Notes in Computer Science ((LNIP,volume 4179))

  • 900 Accesses

Abstract

The nearest neighbor (NN) classifier is well suited for generic object recognition. However, it requires storing the complete training data, and classification time is linear in the amount of data. There are several approaches to improve runtime and/or memory requirements of nearest neighbor methods: Thinning methods select and store only part of the training data for the classifier. Efficient query structures reduce query times. In this paper, we present an experimental comparison and analysis of such methods using the ETH-80 database. We evaluate the following algorithms. Thinning: condensed nearest neighbor, reduced nearest neighbor, Baram’s algorithm, the Baram-RNN hybrid algorithm, Gabriel and GSASH thinning. Query structures: kd-tree and approximate nearest neighbor. For the first four thinning algorithms, we also present an extension to k-NN which allows tuning the trade-off between data reduction and classifier degradation. The experiments show that most of the above methods are well suited for generic object recognition.

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.

Similar content being viewed by others

References

  1. Mattern, F., Denzler, J.: Comparison of appearance based methods for generic object recognition. Pattern Recognition and Image Analysis 14, 255–261 (2004)

    Google Scholar 

  2. Toussaint, G.: Geometric proximity graphs for improving nearest neighbor methods in instance-based learning and data mining. Int. J. of Comp. Geom. & Appl. 15, 101–150 (2005)

    Article  MATH  MathSciNet  Google Scholar 

  3. Clarkson, K.: A randomized algorithm for closest-point queries. SIAM Journal of Computing 17, 830–847 (1988)

    Article  MATH  MathSciNet  Google Scholar 

  4. Dobkin, D., Lipton, R.: Multidimensional searching problems. SIAM Journal of Computing 2, 181–186 (1976)

    Article  MathSciNet  Google Scholar 

  5. Meisner, S.: Point location in arrangements of hyperplanes. Information and Computation 2, 286–303 (1993)

    Article  Google Scholar 

  6. Yao, A., Yao, F.: A general approach to d-dimension geometric queries. In: 17th Symposium on Theory of Computing, pp. 163–168 (1985)

    Google Scholar 

  7. Friedman, J., Bentley, J., Finkel, R.: An algorithm for finding best matches in logarithmic expected time. ACM Transactions on Mathematical Software 3, 209–226 (1977)

    Article  MATH  Google Scholar 

  8. Maneewongvatana, S., Mount, D.: Analysis of approximate nearest neighbor searching with clustered point sets. In: The DIMACS Series in Discrete Mathematics and Theoretical Computer Science, vol. 59, pp. 105–123 (2002)

    Google Scholar 

  9. Arya, S., Mount, D.: Approximate nearest neighbor queries in fixed dimensions. In: Fifth Annual ACM-SIAM Symposium on Discrete Algorithms, pp. 271–280 (1993)

    Google Scholar 

  10. Arya, S., Mount, D., Netanyahu, N., Silverman, R., Wu, A.: An optimal algorithm for approximate nearest neighbor searching. Journal of the ACM 45, 891–923 (1998)

    Article  MATH  MathSciNet  Google Scholar 

  11. Hart, P.E.: The condensed nearest neighbour rule. IEEE Transactions on Information Theory 14, 515–516 (1968)

    Article  Google Scholar 

  12. Gates, W.: The reduced nearest neighbour rule. IEEE Transactions on Information Theory 18, 431–433 (1972)

    Article  Google Scholar 

  13. Baram, Y.: A geometric approach to consistent classification. Pattern Recognition 13, 177–184 (2000)

    Article  Google Scholar 

  14. Olorunleke, O.: Decision Rules for Classification: Classifying Cars into City-Cycle Miles per Gallon Groups. Dep. of Computer Science, University of Saskatchewan, Canada (2003)

    Google Scholar 

  15. Toussaint, G., Bhattacharya, B., Poulsen, R.: The application of voronoi diagrams to non-parametric decision rules. In: 16th Symp. on Comp. Science and Statistics, pp. 97–108 (1984)

    Google Scholar 

  16. Sánchez, J., Pla, F., Ferri, F.: Prototype selection for the nearest neighbor rule through proximity graphs. Pattern Recognition Letters 18, 507–513 (1997)

    Article  Google Scholar 

  17. Mukherjee, K.: Application of the Gabriel graph to instance based learning algorithms. PhD thesis, Simon Fraser University (2004)

    Google Scholar 

  18. Leibe, B., Schiele, B.: Analyzing Appearance and Contour Based Methods for Object Categorization. In: Int. Conf. on Comp. Vision and Pattern Recog., vol. 2, pp. 409–415 (2003)

    Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Editor information

Editors and Affiliations

Rights and permissions

Reprints and permissions

Copyright information

© 2006 Springer-Verlag Berlin Heidelberg

About this paper

Cite this paper

Bajramovic, F., Mattern, F., Butko, N., Denzler, J. (2006). A Comparison of Nearest Neighbor Search Algorithms for Generic Object Recognition. In: Blanc-Talon, J., Philips, W., Popescu, D., Scheunders, P. (eds) Advanced Concepts for Intelligent Vision Systems. ACIVS 2006. Lecture Notes in Computer Science, vol 4179. Springer, Berlin, Heidelberg. https://doi.org/10.1007/11864349_108

Download citation

  • DOI: https://doi.org/10.1007/11864349_108

  • Publisher Name: Springer, Berlin, Heidelberg

  • Print ISBN: 978-3-540-44630-9

  • Online ISBN: 978-3-540-44632-3

  • eBook Packages: Computer ScienceComputer Science (R0)

Publish with us

Policies and ethics