Q-stack aging model for face verification

Permanence of biometric features for face verification remains a largely open research problem. Actual and up-to-date at the time of their creation, extracted features and models relevant to a person´s face may eventually become outdated, leading to a failure in the face verification task. If physical characteristics of the individual change over time, their classification model has to be updated. In this paper, we develop a Q-stack classifier that performs face verification across age progression. Originally, Q-stack classifier has been proposed to use class-independent signal quality measures and baseline classifier scores in order to improve classification. In this paper we demonstrate the application of Q-stack classifier on the task of biometric identity verification using face images and associated metadata quality measure - age. We show that the use of the proposed technique allows for reducing the error rates below those of baseline classifier created at the time of enrolment.