Face Recognition from Depth Images with Convolutional Neural Network

In recent years, the rapid development of face recognition technology has made it a hot research field. Depth image has been widely studied in face recognition due to its advantages of three-dimensional information and light insensitivity. The traditional depth image recognition method mainly focuses on the design of manual features, and it is often difficult to achieve an ideal recognition effect. This paper proposes a Convolutional Neural Network (CNN) structure for face recognition in depth images. And experiments on the RGB-D-T face database show that the proposed CNN structure can significantly improve the face recognition accuracy, compared with traditional face recognition methods, such as LBP, moment invariant and PCA.