Skip to Main Content
The problem of recognizing the partial face expression captured from a viewpoint between -/+45deg arises in many fields of science and engineering, this paper propose the novel method for recognizing the partial face expression by using an 2.5D partial face image. The 2.5D partial face image is captured from a viewpoint between -/+45deg. The 3D virtual expression face is reconstructed from 2.5D partial face image. The facial expression is analyzed in term of the distribution of cross points on the face plane, and then the feature of cross points on the face plane is evaluated by original feature contribution. Finally, the significant feature is used for facial expression recognition by means of multiple similarity method. This proposed method has the advantages: 1) the 2.5D partial face image captured from a viewpoint of face rotation between -45deg and + 45deg can be used for recognition, 2) facial parts extraction like eyes, mouth is not required in this recognition system. The experiments were done for four facial expressions (neutral, anger, surprise and smiling) from 21 persons, based on the assumption that the person has been known. The results show the feasibility of the proposed method.