Face verification across ages using discriminative methods and see 5.0 classifier

Abstract

Identifying a person through face biometric and analysis of Facial image has been drawing interest of researchers in the field of Machine learning and Pattern recognition. Face Recognition Across Ages (FRA) is a challenging task due to aging effects like changes in facial shape and texture. In this paper, an attempt is made to describe a schematic using two different discriminative approaches for feature extraction and a see5.0 classifier for classification purposes. One of the feature finding approaches is based on Gradient Orientation Pyramids (GOP) that includes finding of gradient orientations in Gaussian pyramids, the later one is based on Local Binary Patterns (LBP) calculated at each stage of Gaussian pyramid decomposition. These we have used FG-NET database and accuracies of both the approaches are compared.