Age-invariant face recognition using a feature progressing model

Face recognition is to identify a person from the person's face image. Age invariant face recognition is to achieve face recognition while the person's face image may be taken at any age. To tackle this problem, two kinds of methods are often adopted, namely, usage of a facial aging model or elimination of age effects. In this paper, we propose a feature progressing model and use it for age invariant face recognition. The proposed method is evaluated on 2 databases of face images with age variance, namely, FG-NET[1], MORPH[2], and compared with some state-of-art methods. The results demonstrated the advantages of the new model.