DeepFaceAR: Deep Face Recognition and Displaying Personal Information via Augmented Reality
Paper ID : 1011-MVIP2020
Amin Golnari *1, Hossein Khosravi2, Saeid Sanei3
2dept. Electrical and Robotics, Shahrood University of Technology
3School of Science and Technology, Nottingham Trent University
Biometric recognition is a popular problem in machine vision. Deep neural networks have been recently introduced as a solution to this problem. Therefore, in this article, DeepFaceAR combines a convolutional neural network with augmented reality for representation of individuals information. In this approach, we used a dataset containing 100 facial images of the faculty members of Shahrood University of Technology and we were able to obtain an accuracy of 99.45% for their recognition. In the augmented reality section also provides graphical image targets for each person. When a person is identified by the deep neural network, the target image provided for augmented reality is displayed.
Deep Learning, Augmented Reality, AR, CNNs, Face Recognition
Status : Paper Accepted (Oral Presentation)