Abstract: A novel method of calculating the rotation of face is proposed based on the 3D
facial depth information. Facial feature points are extracted by the depth map and its
corresponding gray image according to differential geometry theory, curvature algorithm and
intensity features generated by face data. And then the three pose angles of the face in the space
will be calculated. Experimental results show that the proposed algorithm can achieve an accurate
estimation of face rotation with the variation of pose and provide a priori knowledge for face
recognition and facial expression analysis.

Key words: depth map, facial feature location, facial surface, 3D curvature, estimation of
rotation