Skip to Main Content
This paper proposes a novel method to detect faces in color images with a wide range of variations in head poses, face tilting, facial expressions and body interferences. The proposed algorithm can locates not only the face regions, but also some important facial components such as month corners and eyes. The experimental results show that the proposed method exhibits satisfied performance in both well-controlled and real-world environments. The performance is also compared with that of a face detector developed by CMU to demonstrate the major strength of the proposed algorithm.