Integration of facial position estimation and person identification for face authentication

Abstract

AbstractIn the automatic authentication of face images, the estimation of the facial position is the most important process. When the facial position is estimated on the basis of the facial shape model, the accuracy is not guaranteed unless the optimal model is used. In the solution of this problem, there is a trade‐off relation between the estimation accuracy and the processing time. To deal with this point, this paper proposes a method which integrates the processes for estimation of the facial position, the facial size, and the individual model. The position and the size are estimated by iterative processing based on the coarse–fine procedure and scale conversion. For the estimation of the individual model, a model database based on the hierarchy by clustering is used. In the proposed estimation process, the cluster containing the optimal model is successively selected in accordance with the hierarchical structure, with the progress of the estimation of the position and the size. The finally selected individual model can be considered as the result of person identification. We constructed a face authentication system based on the proposed method and verified its effectiveness by an evaluation experiment. © 2007 Wiley Periodicals, Inc. Syst Comp Jpn, 38(5): 43–58, 2007; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/scj.20489