Study on Automatic Age Estimation and Restoration for Verification of Human Faces

While recognition of most facial variations, like identity, expression, and gender, has been extensively studied, automatic age estimation has seldom been explored. In distinction to alternative facial variations, aging variation presents many unique characteristics that build age estimation a difficult task. This paper proposes associate degree automatic age estimation methodology named AGES (AGing pattern Subspace). The fundamental plan is to model the aging pattern that is outlined because the sequence of a selected individual’s face pictures sorted in time order, by constructing a representative topological space. The right aging pattern for a antecedently unseen face image is decided by the projection within the topological space which will reconstruct the face image with minimum reconstruction error, whereas the position of the face image in this aging pattern can then indicate its age.