Development of Face Recognition System Using Multi-views Database

Xiaojuan Liu, Jianwu Dang, Fan Yang, Michel Paindavoine

Abstract: This paper presents a new multi-views face database and some development results on face recognition tasks based on it. Our objective is to find an efficient and simple algorithm for performing face recognition in real time. The image acquisition system is composed of five standard cameras, which together can simultaneously take five pictures of a human face from different perspectives. We address the problem of face recognition using independent component analysis (ICA). We explore the issues of subspace selection, algorithm comparison, and multi-views face recognition performance. In order to make full use of the multi-views property, we also propose a strategy of majority voting among the five views, which can improve the recognition rate. Experimental results show that ICA is a promising method among the many possible face recognition methods, and that the ICA algorithm with majority-voting is currently the best choice for our purposes.