Face Verification By Sharing Knowledge From Different Subjects

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 approximations in the related literature.