A geometrical-model-based face recognition

This paper presents a novel two-stage face recognition method. The first stage implements a newly proposed feature, called Local Vector Pattern (LVP), and with a weighting mechanism, LVP is used to compute the distance between an input face image and each enrolled face image, so that M possible face candidates can be decided. Then, the second stage uses a novel feature-point Bilateral Recognition (BR) approach to produce the final face recognition result from the M candidates. Bilateral recognition contains both forward and backward recognition, and each recognition uses the feature points detected in its reference image, and searches their individual corresponding feature points in its matched image. The geometrical models of the detected feature points and their matched ones are respectively constructed and compared. By summing up the scores of both forward and backward recognition, a bilateral recognition score is obtained and is used to produce the final recognition result. Experiments on the famous Feret face databases show that the proposed algorithm produce an excellent recognition result and performs much better than two other well-known face recognition methods.