Design, Implementation and Evaluation of Hardware Vision Systems Dedicated to Real-Time Face Recognition

Overview & Synopsis

Human face recognition is an active area of research spanning several disciplines such as image processing, pattern recognition, and computer vision. Most researches have concentrated on the algorithms of segmentation, feature extraction, and recognition of human faces, which are generally realized by software implementation on standard computers. However, many applications of human face recognition such as human-computer interfaces, model-based video coding, and security control (Kobayashi, 2001, Yeh & Lee, 1999) need to be high-speed and real-time, for example, passing through customs quickly while ensuring security. For the last years, our laboratory has focused on face processing and obtained interesting results concerning face tracking and recognition by implementing original dedicated hardware systems. Our aim is to implement on embedded systems efficient models of unconstrained face tracking and identity verification in arbitrary scenes. The main goal of these various systems is to provide efficient robustness algorithms that only require moderated computation in order 1) to obtain high success rates of face tracking and identity verification and 2) to cope with the drastic real-time constraints. The goal of this chapter is to describe three different hardware platforms dedicated to face recognition. Each of them has been designed, implemented and evaluated in our laboratory.