Face View Recognition and Facial Feature Extraction

Issues of friendly human-machine interfaces have driven various researches, such as face animation, face recognition, and facial expression recognition in the last decade. The success of these applications relies on the accuracy of the detected facial features. Despite that facial features may be extracted automatically by some methods, however, the effective results were obtained in certain restricted conditions, such as face images acquired in the front view. This study proposed a facial feature extraction method in different views. A contour descriptor is adopted to describe the shape of faces. Then, the views were automatically recognized using a radial basis function neural network. According to the recognition result, the respective facial feature extraction was applied. Experimental results show that the proposed method can recognize face views (front view, half profile view, and profile view) and extract their corresponding feature points correctly.