Geometric feature based facial expression recognition using multiclass support vector machines

In this paper, a novel geometric features extraction method for facial expression recognition is proposed. ASM automatic fiducial point location algorithm is firstly applied to a facial expression image, and then calculating the Euclidean distances between the center of gravity coordinate and the annotated fiducial points´ coordinates of the face image. In order to extract the discriminate deformable geometric information, the system extracts the geometric deformation difference features between a person´s neural expression and the other seven basic expressions. A multiclass support vector machine (SVM) classifier is used to recognize the facial expressions. Experiments indicate that our proposed method can obtain good classification accuracy.