In this paper, a novel glasses-face recognition approach, eyeglasses eigenfaces, is proposed to recognize glasses-face, which treats eyeglasses as a feature of facial image. It overcomes the choke point of removing eyeglasses which used in previous glasses-face recognition methods. Considering the instability of eyeglasses as a facial feature, here we make use of 3D face synthesis method based on genetic algorithm to reconstruct virtual samples, enrich the sample library. It not only can be used in different pose, illumination and expression, but also provide a new thought-way for occlusion problem in face recognition. On CAS-PEAL face databases, our experimental results demonstrate that eyeglasses eigenfaces perform well.
展开▼
