Face based Automatic Kinship Veri cation

Abstract

Face based automatic kinship veri cation is a novel challenging research problem in computer vision. It performs the automatic examining of the facial attributes and expecting whether two persons have a biological kin relation or not. The focus is on providing novel solutions for challenges of family veri cation from faces with an e cient system with the aim of providing enhancement to the accuracy of kinship veri cation. In our work, we analyzed the facial kinship veri cation systems in two modes the unimodal and multi-modal system. The feature extraction is a crucial step in the kinship recognition system. For this reason, we proposed two e cient feature learning extraction algorithms called discrete cosine transform network (DCTNet) and Context-Aware Local Binary Feature Learning (CA-LBFL). Various databases are used and extensive experiments are carried out in order to validate our proposed methods and developed methods. Besides, the experimental results demonstrated that the proposed methods achieved competitive results compared with other state-of-the-art.