3D Facial Expression Recognition Method Based on Local Curvature

In order to improve the recognition rate and recognition speed of 3D facial expression recognition, the 3D facial expression recognition method is proposed by local curvature in this paper. First, the tip point of the nose is extracted by the face center profile, with the tip point of the nose as the reference point, searching for search windows of other feature points and automatically extract feature points through local curvature in its window. These feature points are composed into feature vector. Finally, K-means algorithm is adopted to expression classification. The theoretical analysis and experimental results both show that this method has greatly improve the recognition rate and recognition speed of 3D facial expression recognition.