Skip to Main Content
This paper presents a feature-based approach for fast face recognition. A novel shape-based automatic reference control point and feature extraction technique is proposed for face representation, whereby the difference between two faces is measured by a set of extracted features, and 3-D features from a set of 2-D images are used for face template registration. Unlike holistic face recognition algorithms, the feature-based algorithm is relatively robust to variations of face expressions, illumination, and pose, due to invariance of its facial feature vector. The theoretical performance analysis of the proposed technique was provided by a probabilistic and statistical approach. The proposed approach is shown to achieve promising performance for face recognition using several subsets of face recognition databases.