An efficiently method for recognizing 3D face based on constructed character

In order to eliminate the effect of the facial expression, and also improve the accuracy of the three dimensional (3D) face recognition, a new 3D face recognition method is proposed in this paper. First, segmenting the key regions and curves which are free from the facial and fully characterize the face; second, resampling curves and regions to uniform the data precision; third, processing data to extract the feature; finally, matching the features of curves and regions to identify 3D face. The Experiment results in actual 3D face database demonstrate the availability and effectiveness of the presented method.