Face recognition using convolutional neural networks

Abstract

Face Recognition is a currently developing technology with multiple real-life applications. Deep learning, in particular Convolutional Neural Networks, has achieved promising results in Face Recognition recently. The goal of this project is to implement a complete Face Recognition system based on state-of-the art Convolutional Neural Networks, and to test it on effected images in unconstrained environments. Relevant facial features are extracted and used to determine if pairs of face images belong to the same individual or not. These features allow to compare faces between them in an efficient way. The system is trained using two different datasets, Faces94 and FaceScrub, resulting in two models. The "Labeled Faces in the Wild" benchmark is used for testing and evaluating the performance of the two models. A comparison with classical methods is also done.