Explainable Face Recognition Framework and Qualitative Fairness Assessment

Explainable Face Recognition (FR) has now become increasingly im- portant in our lives, like unlocking smartphones or automated border checkpoints. Because of this potential for pervasiveness in human life, it is important to understand why two face images are matched or not matched by a given face recognition system. The objective of this project is to reproduce the results of a recent paper on Explainable Face Recog- nition and evaluate it also on the RFW dataset[3] to assess the presence or absence of biases.