FACE VERIFICATION BY SHARING KNOWLEDGE FROM DIFFERENT SUBJECTS

David Masip, Àgata Lapedriza, Jordi Vitrià

Abstract

In face verification problems the number of training samples from each class is usually reduced, making difficult the estimation of the classifier parameters. In this paper we propose a new method for face verification where we simultaneously train different face verification tasks, sharing the model parameter space. We use a multi-task extended logistic regression classifier to perform the classification. Our approach allows to share information from different classification tasks (transfer knowledge), mitigating the effects of the reduced sample size problem. Our experiments performed using the publicly available AR Face Database, show lower error rates when multiple tasks are jointly trained sharing information, which confirms the theoretical approx- imations in the related literature.

Download


Paper Citation


in Harvard Style

Masip D., Lapedriza À. and Vitrià J. (2007). FACE VERIFICATION BY SHARING KNOWLEDGE FROM DIFFERENT SUBJECTS . In Proceedings of the Second International Conference on Computer Vision Theory and Applications - Volume 2: VISAPP, ISBN 978-972-8865-74-0, pages 286-289. DOI: 10.5220/0002058902860289


in Bibtex Style

@conference{visapp07,
author={David Masip and Àgata Lapedriza and Jordi Vitrià},
title={FACE VERIFICATION BY SHARING KNOWLEDGE FROM DIFFERENT SUBJECTS},
booktitle={Proceedings of the Second International Conference on Computer Vision Theory and Applications - Volume 2: VISAPP,},
year={2007},
pages={286-289},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0002058902860289},
isbn={978-972-8865-74-0},
}


in EndNote Style

TY - CONF
JO - Proceedings of the Second International Conference on Computer Vision Theory and Applications - Volume 2: VISAPP,
TI - FACE VERIFICATION BY SHARING KNOWLEDGE FROM DIFFERENT SUBJECTS
SN - 978-972-8865-74-0
AU - Masip D.
AU - Lapedriza À.
AU - Vitrià J.
PY - 2007
SP - 286
EP - 289
DO - 10.5220/0002058902860289