Feature based Age Prediction for Face Recognition

Abstract

Face recognition is currently highly researched area of computer vision and pattern recognition. Age changes cause major variations in the appearance of human faces. In this paper, we propose a new aging scheme, which is ultimately to be used in face recognition applications. It will provide to reduce the memory space and processing times. This proposal requires only one input image of a subject and produces an age progressed image of the person at a particular target age. To implement this training-based approach, over 300 high quality facial images ranging from 18 years to 70 years in age were collected from the Internet. An age prediction function is a relationship between a facial image and its corresponding age. Given a single image of the subject and a target age, we construct a new image of the subject to reflect the new age. Such an image will be useful to predict a person’s aged face for use in face recognition systems.