Research on a facial feature points detection method based on skin color model

Abstract

In recent years, facial feature points detection has become a popular research topic in the field of computer vision. The traditional facial feature points detection methods are defective, because these methods are affected by expression change, illumination change and other factors. In this paper, the skin color model is combined with gray information to detect and locate the facial feature points. Firstly, the skin color model is used to segment the image and extract the facial feature region. After the mathematical morphology processing, the gray integral projection method is used to detect and locate the feature points accurately. Compared to other traditional detected method, this method solves the negative effects of the factors such as expression change and illumination change on the feature points detection. It can also show high precision and strong robustness in real-time video detection system.