Comparative study of invariant descriptors for face recognition
Résumé
We present in this paper a comparative study of invariant descriptors for face recognition. Despite of their multiple advantages (pose and scale invariant characterization), few methods from the state of the art exploit this kind of features. A support vector machine is used for the training and recognition steps. Experimental results show the efficiency of such descriptors on the AR face database. With the proposed method, face recognition is correctly achieved in 97.4% cases