This project investigates the recent advances in research on face detection and recognition to design an automatic student attendance taking prototype. Designing such prototype requires performing two major tasks, namely face detection and face recognition. The proposed solution consists of three main modules that are, Face Detection module, Face Recognition module and Postprocessing module. The face detection module uses a tuned version of the Viola-Jones algorithm due to its effectiveness and high speed. The face recognition module modifies and adapts a pre-trained Convolutional Neural Network (VGG-face). This choice is due to the high accuracy of ConvNets. The postprocessing module consists of a rule-based algorithm that validates the output of the recognition module. The results of the experiments we have conducted showed the effectiveness of the proposed solution since attendance lists resulting processing 600 class images were all correct.