Four Principles of Explainable AI as Applied to Biometrics and Facial Forensic Algorithms

TL;DR

This paper proposes four principles of explainable AI tailored for biometric and facial recognition systems, illustrated through case studies highlighting challenges in developing explainable algorithms.

Contribution

It introduces four principles of explainable AI specifically applied to face recognition and biometrics, addressing the gap between algorithm accuracy and societal trust.

Findings

Four principles of explainable AI for biometrics

Case studies demonstrating explanation challenges

Insights into developing trustworthy biometric algorithms

Abstract

Traditionally, researchers in automatic face recognition and biometric technologies have focused on developing accurate algorithms. With this technology being integrated into operational systems, engineers and scientists are being asked, do these systems meet societal norms? The origin of this line of inquiry is `trust' of artificial intelligence (AI) systems. In this paper, we concentrate on adapting explainable AI to face recognition and biometrics, and we present four principles of explainable AI to face recognition and biometrics. The principles are illustrated by $\it{four}$ case studies, which show the challenges and issues in developing algorithms that can produce explanations.