3D Facial Expression Recognition on Curvature Local Binary Patterns

Abstract

This paper proposed a new method Based on curvature Based LBP feature to recognize 3D facial expression automatically. 3D facial expression images are described by means of four images which gray level are the value of curvature-Based descriptors (principal curvatures k1, k2, mean curvature, shape index) and then encoded by LBP. To efficiently optimize the performance, Chi-square distance is employed for classification. Finally, experimental result achieved on the Bosphorus database illustrates that the curvature-Based LBP (CLBP) has performs better than other features and also shows these features are significant for 3D facial expression recognition.