Combination of Tangent Vectors and Local Representations for Handwritten Digit Recognition

  • Daniel Keysers
  • Roberto Paredes
  • Hermann Ney
  • Enrique Vidal
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 2396)


Statistical classification using tangent vectors and classification based on local features are two successful methods for various image recognition problems. These two approaches tolerate global and local transformations of the images, respectively. Tangent vectors can be used to obtain global invariance with respect to small affine transformations and line thickness, for example. On the other hand, a classifier based on local representations admits the distortion of parts of the image. From these properties, a combination of the two approaches seems very likely to improve on the results of the individual approaches. In this paper, we show the benefits of this combination by applying it to the well known USPS handwritten digits recognition task. An error rate of 2.0% is obtained, which is the best result published so far for this dataset.


Local Feature Tangent Vector Near Neighbor Local Representation Relevance Vector Machine 
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.


  1. 1.
    R. Duda, P. Hart, and D. Stork. Pattern Classification. John Wiley & Sons, New York, 2nd edition, 2001.zbMATHGoogle Scholar
  2. 2.
    P. Simard, Y. Le Cun, and J. Denker. Efficient Pattern Recognition Using a New Transformation Distance. In S. Hanson, J. Cowan, and C. Giles, editors, Advances in Neural Information Processing Systems 5. Morgan Kaufmann, pages 50–58, 1993.Google Scholar
  3. 3.
    D. Keysers, W. Macherey, J. Dahmen, and H. Ney. Learning of Variability for Invariant Statistical Pattern Recognition. In ECML 2001, 12th European Conference on Machine Learning, volume 2167 of Lecture Notes in Computer Science, Springer, Freiburg, Germany, pages 263–275, September 2001.CrossRefGoogle Scholar
  4. 4.
    J. Dahmen, D. Keysers, and H. Ney. Combined Classification of Handwritten Digits using the’ Virtual Test Sample Method’. In MCS 2001, 2nd International Workshop on Multiple Classifier Systems, volume 2096 of Lecture Notes in Computer Science, Springer, Cambridge, UK, pages 109–118, May 2001.CrossRefGoogle Scholar
  5. 5.
    C. Schmid and R. Mohr. Local grayvalue invariants for image retrieval. IEEE Transactions on Pattern Analysis and Machine Intelligence, 19(5):530–535, 1997.CrossRefGoogle Scholar
  6. 6.
    C. Shyu, C. Brodley, A. Kak, A. Kosaka, A. Aisen, and L. Broderick. Local versus Global Features for Content-Based Image Retrieval. In Proc. of the IEEE Workshop on Content-Based Access of Image and Video Libraries, Santa Barbara, CA, pages 30–34, June 1998.Google Scholar
  7. 7.
    R. Deriche and G. Giraudon. A Computational Approach to Corner and Vertex Detection. Int. Journal of Computer Vision, 10:101–124, 1993.CrossRefGoogle Scholar
  8. 8.
    R. Paredes, J. Perez-Cortes, A. Juan, and E. Vidal. Local Representations and a Direct Voting Scheme for Face Recognition. In Workshop on Pattern Recognition in Information Systems, Setúbal, Portugal, pages 71–79, July 2001.Google Scholar
  9. 9.
    S. Arya, D. Mount, N. Netanyahu, R. Silverman, and A. Wu. An optimal algorithm for approximate nearest neighbor searching fixed dimensions. Journal of the ACM, 45:891–923, 1998.zbMATHCrossRefMathSciNetGoogle Scholar
  10. 10.
    J. Kittler, M. Hatef, and R. Duin. Combining Classifiers. In Proceedings 13th International Conference on Pattern Recognition, Vienna, Austria, pages 897–901, August 1996.Google Scholar
  11. 11.
    M. Tipping. The Relevance Vector Machine. In S. Solla, T. Leen, and K. Müller, editors, Advances in Neural Information Processing Systems 12. MIT Press, pages 332–388, 2000.Google Scholar
  12. 12.
    P. Simard, Y. Le Cun, J. Denker, and B. Victorri. Transformation Invariance in Pattern Recognition — Tangent Distance and Tangent Propagation. In Neural Networks: Tricks of the Trade, volume 1524 of Lecture Notes in Computer Science, Springer, Heidelberg, pages 239–274, 1998.CrossRefGoogle Scholar
  13. 13.
    B. Schölkopf, P. Simard, A. Smola, and V. Vapnik. Prior Knowledge in Support Vector Kernels. In M. Jordan, M. Kearns, and S. Solla, editors, Advances in Neural Inf. Proc. Systems, volume 10. MIT Press, pages 640–646, 1998.Google Scholar
  14. 14.
    D. Keysers, J. Dahmen, T. Theiner, and H. Ney. Experiments with an Extended Tangent Distance. In Proceedings 15th International Conference on Pattern Recognition, volume 2, Barcelona, Spain, pages 38–42, September 2000.Google Scholar
  15. 15.
    J. Dahmen, D. Keysers, H. Ney, and M. O. Güld. Statistical Image Object Recognition using Mixture Densities. Journal of Mathematical Imaging and Vision, 14(3):285–296, May 2001.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2002

Authors and Affiliations

  • Daniel Keysers
    • 1
  • Roberto Paredes
    • 2
  • Hermann Ney
    • 1
  • Enrique Vidal
    • 2
  1. 1.Lehrstuhl für Informatik VI - Computer Science DepartmentRWTH Aachen - University of TechnologyAachenGermany
  2. 2.Instituto Tecnológico de InformáticaUniversidad Politécnica de ValenciaValenciaSpain

Personalised recommendations