Title:Deep learning for identification and face, gender, expression recognition under constraints

Abstract:Biometric recognition based on the full face is an extensive research area. However, using only partially visible faces, such as in the case of veiled-persons, is a challenging task. Deep convolutional neural network (CNN) is used in this work to extract the features from veiled-person face images. We found that the sixth and the seventh fully connected layers, FC6 and FC7 respectively, in the structure of the VGG19 network provide robust features with each of these two layers containing 4096 features. The main objective of this work is to test the ability of deep learning based automated computer system to identify not only persons, but also to perform recognition of gender, age, and facial expressions such as eye smile. Our experimental results indicate that we obtain high accuracy for all the tasks. The best recorded accuracy values are up to 99.95% for identifying persons, 99.9% for gender recognition, 99.9% for age recognition and 80.9% for facial expression (eye smile) recognition.