3D face recognition system based on feature analysis and support vector machine
Jiann-Der Lee, Chen-Hui Kuo, Chen‐Min Hsu · 2004
In this paper, a novel 3D face recognition system based on feature analysis and support vector machine (SVM) is proposed. The first stage of this approach is to normalize the altitude and angle of 3D facial data to remove the distortion resulted from the head pose under arbitrary rotation. Next, the chain code method is employed for feature extraction in several selected facial regions. With the aids of the factor analysis techniques, the number of features is effectively reduced from 26 to 10, which decreased massive computation cost and make the whole system more efficiently. From the experimental results, it is observed that the correction rate using the recognition scheme based on SVM achieves up to 98%, which proves the superior performance of this system.