A face detection method via combination of skin color detection and visual attention theory

Since the face detection technology risen. It has been an important research subject of computer vision and pattern recognition. A rapid and accurate method of face detection is the important premise of guarantee for face recognition and pedestrian tracking research fields. This paper proposes a face detection method via skin color detection and visual attention theory. Firstly, uses the Saliency Detection via Graph-Based Manifold Ranking to get the picture's salient region, Secondly, in the salient region, uses the skin color detection method which based on YCBCR color space to detect the face candidate regions, Finally, after a discrimination confirm the face location. The experimental results show that the method can quickly and accurately detect the human face, robust to illumination change as well.