Similarity Function Learning with Data Uncertainty

Julien Bohné, Sylvain Colin, Stéphane Gentric, Massimiliano Pontil


Similarity functions are at the core of many pattern recognition applications. Standard approaches use feature vectors extracted from a pair of images to compute their degree of similarity. Often feature vectors are noisy and a direct application of standard similarly learning methods may result in unsatisfactory performance. However, information on statistical properties of the feature extraction process may be available, such as the covariance matrix of the observation noise. In this paper, we present a method which exploits this information to improve the process of learning a similarity function. Our approach is composed of an unsupervised dimensionality reduction stage and the similarity function itself. Uncertainty is taken into account throughout the whole processing pipeline during both training and testing. Our method is based on probabilistic models of the data and we propose EM algorithms to estimate their parameters. In experiments we show that the use of uncertainty significantly outperform other standard similarity function learning methods on challenging tasks.


  1. Belhumeur, P. N., ao P. Hespanha, J., and Kriegman, D. J. (1997). Eigenfaces vs. fisherfaces: Recognition using class specific linear projection.IEEE Transactions on Pattern Analysis and Machine Intelligence, pages 711-720.
  2. Bi, J. and Zhang, T. (2004). Support vector classification with input data uncertainty. In NIPS, pages 1651- 1659.
  3. Blanz, V., Grother, P., Phillips, J. P., and Vetter, T. (2005). Face recognition based on frontal views generated from non-frontal images. In CVPR, pages 454-461.
  4. Bohné, J., Ying, Y., Gentric, S., and Pontil, M. (2014). Large margin local metric learning. In ECCV.
  5. Cao, X., Wipf, D., Wen, F., and Duan, G. (2013). A practical transfer learning algorithm for face verification. In ICCV.
  6. Chen, D., Cao, X., Wang, L., Wen, G., and Sun, J. (2012). Bayesian face revisited: a joint formulation. In ECCV.
  7. Cormode, G. and McGregor, A. (2008). Approximation algorithms for clustering uncertain data. In PODS.
  8. Davis, J. V., Kulis, B., Jain, P., Sra, S., and Dhillon, I. S. (2007). Information-theoretic metric learning. In ICML, pages 209-216.
  9. Guillaumin, M., Verbeek, J., and Schmid, C. (2009). Is that you? metric learning approaches for face identification. In ICCV, pages 498-505.
  10. Huang, G. B., Ramesh, M., Berg, T., and Learned-Miller, E. (2007). Labeled faces in the wild: A database for studying face recognition in unconstrained environments. Technical Report 07-49, University of Massachusetts, Amherst.
  11. Köstinger, M., Hirzer, M., Wohlhart, P., Roth, P. M., and Bischof, H. (2012). Large scale metric learning from equivalence constraints. In CVPR, pages 2288-2295.
  12. Kriegel, H.-P. and Pfeifle, M. (2005). Hierarchical densitybased clustering of uncertain data. In ICDM.
  13. Li, S. Z. and Jain, A. K. (2011). Handbook of Face Recognition 2nd ed. Springer.
  14. Prince, S. J. and Elder, J. H. (2007). Probabilistic linear discriminant analysis for inferences about identity. In ICCV.
  15. Ren, J., Lee, S. D., Chen, X., Kao, B., Cheng, R., and Cheung, D. W.-L. (2009). Naive bayes classification of uncertain data. In ICDM.
  16. Shivaswamy, P. K., Bhattacharyya, C., and Smola, A. J. (2006). Second order cone programming approaches for handling missing and uncertain data. Journal of Machine Learning Research, 7:1283-1314.
  17. Sun, Y., Chen, Y., Wang, X., and Tang, X. (2014a). Deep learning face representation by joint identificationverification. InNIPS.
  18. Sun, Y., Wang, X., and Tang, X. (2014b). Deep learning face representation from predicting 10,000 classes. In CVPR.
  19. Tipping, M. E. and Bishop, C. M. (1999). Probabilistic principal component analysis. Journal of the Royal Statistical Society, Series B, 61:611-622.
  20. Tsang, S., Kao, B., Yip, K. Y., Ho, W.-S., and Lee, S. D. (2011). Decision trees for uncertain data. IEEE Transactions on Knowledge and Data Engineering, 23:64- 78.
  21. Weinberger, K. and Saul, L. (2009). Distance metric learning for large margin nearest neighbor classification. Journal of Machine Learning Research, 10:207-244.

Paper Citation

in Harvard Style

Bohné J., Colin S., Gentric S. and Pontil M. (2016). Similarity Function Learning with Data Uncertainty . In Proceedings of the 5th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM, ISBN 978-989-758-173-1, pages 131-140. DOI: 10.5220/0005648601310140

in Bibtex Style

author={Julien Bohné and Sylvain Colin and Stéphane Gentric and Massimiliano Pontil},
title={Similarity Function Learning with Data Uncertainty},
booktitle={Proceedings of the 5th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM,},

in EndNote Style

JO - Proceedings of the 5th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM,
TI - Similarity Function Learning with Data Uncertainty
SN - 978-989-758-173-1
AU - Bohné J.
AU - Colin S.
AU - Gentric S.
AU - Pontil M.
PY - 2016
SP - 131
EP - 140
DO - 10.5220/0005648601310140