I. Introduction
Nowadays, automatically recognizing the identity of a person is of paramount importance. Face recognition is one biometric modality that is widely used in personal devices, but also in legal and commercial applications, with high acceptability, collectability and universality [1], [2]. Advances in imaging technologies have been leading to a revolution in face recognition systems over the last decade. The recent availability of new imaging sensors opens a new range of possibilities also for image-based personal recognition systems, notably allowing a boost in the performance of many face recognition systems and applications [3]. Light field imaging technologies have recently come into prominence as they record not only the intensity of light on a specific 2D plane position but rather the intensity of the light rays for any direction in space [4], [5]. Preliminary works have shown the effectiveness of the supplementary information captured by light field cameras for face recognition applications [6], [7], even when considering one single shot and without using additional optics.