Authors:
Julien Bohné
1
;
Sylvain Colin
2
;
Stéphane Gentric
2
and
Massimiliano Pontil
3
Affiliations:
1
Safran Morpho and University College London, France
;
2
Safran Morpho, France
;
3
University College London, United Kingdom
Keyword(s):
Similarity Function, Uncertain Data, Missing Data, Face Recognition.
Related
Ontology
Subjects/Areas/Topics:
Applications
;
Bayesian Models
;
Biomedical Engineering
;
Biomedical Signal Processing
;
Biometrics
;
Biometrics and Pattern Recognition
;
Missing Data
;
Multimedia
;
Multimedia Signal Processing
;
Pattern Recognition
;
Similarity and Distance Learning
;
Telecommunications
;
Theory and Methods
Abstract:
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 sign
ificantly outperform other standard similarity function learning methods on challenging tasks.
(More)