A technique for face verification across age progression with large age gap

Advancements in the field of biometrics have led to the development of face identification and verification systems. The problem of face verification faces many challenges; however, the challenge of progressing age for verification of human faces remains to be the most challenging of all. The work in the field of face verification across age progression is limited and the field requires more attention and research. This work proposes a technique for face verification which can handle large age gaps. The proposed technique extracts Center Symmetric Local Binary Patterns (CSLBP) to represent the features of the face image. CSLBP has been widely used as a texture descriptor in face recognition. The performance of CSLBP features when combined with weighted K-Nearest Neighbor classifier is best when compared with Support Vector Machine and Tree based classifiers. This work studies the performance of the proposed method on increasing age gaps. Usually, the performance of any face verification system deteriorates or stagnates as the age gap increases; however, the performance of the proposed method improves as the age gap increases. The experiments are performed on MORPH database.