DocumentCode :
716137
Title :
Sparse support faces
Author :
Biggio, Battista ; Melis, Marco ; Fumera, Giorgio ; Roli, Fabio
Author_Institution :
Dept. of Electr. & Electron. Eng., Univ. of Cagliari, Cagliari, Italy
fYear :
2015
fDate :
19-22 May 2015
Firstpage :
208
Lastpage :
213
Abstract :
Many modern face verification algorithms use a small set of reference templates to save memory and computational resources. However, both the reference templates and the combination of the corresponding matching scores are heuristically chosen. In this paper, we propose a well-principled approach, named sparse support faces, that can outperform state-of-the-art methods both in terms of recognition accuracy and number of required face templates, by jointly learning an optimal combination of matching scores and the corresponding subset of face templates. For each client, our method learns a support vector machine using the given matching algorithm as the kernel function, and determines a set of reference templates, that we call support faces, corresponding to its support vectors. It then drastically reduces the number of templates, without affecting recognition accuracy, by learning a set of virtual faces as well-principled transformations of the initial support faces. The use of a very small set of support face templates makes the decisions of our approach also easily interpretable for designers and end users of the face verification system.
Keywords :
face recognition; image matching; learning (artificial intelligence); support vector machines; face recognition; face verification algorithms; kernel function; matching scores; reference templates; sparse support faces; support face templates; support vector machine; virtual face learning; well-principled approach; Accuracy; Face recognition; Integrated circuits; Kernel; Support vector machines; Training; Training data;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Biometrics (ICB), 2015 International Conference on
Conference_Location :
Phuket
Type :
conf
DOI :
10.1109/ICB.2015.7139053
Filename :
7139053
Link To Document :
بازگشت