Designing Intuitive and Effective Dynamic Facial Authentication: Machine Interaction with Human Factors

Abstract

Abstract: The objective of this paper is to propose a spoof-free Face Liveness Detection system with an active approach that uses the challenge-response methodology which detects the motions and gestures of the user, thereby analyzing and recognizing a real face from a photo. With cybercrimes on the rise each day, identity thefts being one of them- especially in an unsupervised authentication system, has given rise to serious security concerns. Face liveness detection assures the user's actual presence and validates their identities. Along with that, it prevents fraudulent reproduction of face biometrics to spoof the system and thus effectively protects the rights and data of the legitimate user. A review of the many scholarly pieces of the literature revealed works of some of the existing systems that use active or passive techniques to test liveness though there is still a need for a more robust approach. Predicted results of the Human-Computer Interaction (HCI) based approach demonstrated high accuracy in detecting the liveness of the user which effectively reduced system vulnerability to face-spoof attacks. The aim is to create a system that is highly accurate, efficient, and can be integrated with existing systems.