Face anti-spoofing with image quality assessment

Face recognition-based authentication techniques can be easily spoofed using various types of attack. Consistent counter-measures need to meet certain requirements, mainly regarding reliable robustness and low complexity. In this paper, we aim to find the best compromise between these two criteria, as we propose an anti-spoofing solution based on Image Quality Assessment (IQA) to distinguish between genuine and fake face-appearances. Our solution was tested on a publicly available database, and proved to outperform state-of-the-art approaches.