New deep learning based face recognition attendance system is proposed in the paper. The model is composed of several steps developed using currently most advanced techniques: convolutional neural network (CNN) cascade for face detection and CNN for generating face embeddings. Primary goal of the research was the practical employment of this state-of-the-art deep learning approaches for face recognition tasks. The overall accuracy is 95.02% on a small dataset of original face images in the real-time environment.
Scientific paper published on International Conference SISY2017