With the improvement in the quality and availability of capture devices such as smartphones, tablets, and webcams, authentication applications are available to the general public and can be used in uncontrolled environments and without training, but they can also produce low-quality images due to factors such as poor illumination, movement, non-frontal pose, and facial occlusions.

Given these challenges, the ability to analyze the quality of a facial image is essential for any face-processing task, including recognition, presentation attack detection (PAD), and age estimation, amongst others, all of which are based on deep-learning neural network models. To address these issues, we have built an AI-powered suite of facial attributes and image quality assessment tools (FAIQA), which can be integrated into applications with the objective of improving the quality of the capture.