The facial features analysis method based on human star-structured model

This paper proposes a face detection and facial feature points location algorithm based on the human starstructured model. The head model of human star-structured model based on deformable part models (DPM) can locate the pedestrian’s head area. Haar feature and Adaboost cascaded classifier are used to detect face on the head area. Then, local binary features from face are extracted based on random forest. Global linear regression model are learned from all the local binary features and then the facial feature points are located. The method for face detection and feature points location based on human star-structured model reduces the time from scale and iterate through the images, therefore reduces the time of face detecting and feature points locating.