Research on Face Detection and Recognition with Face Mask Based on FaceNet
Hongxia Wang, Yuting Gu, Hui Li · 2022
According to the requirements of face recognition with masks in epidemic environment, this paper analyzes the difficulties of face recognition with masks, and puts forward a scheme combining face detection, data enhancement and face recognition. Firstly, face detection is performed by MtCNN algorithm; Then, data enhancement technology is used to enhance the complexity of sample data sets and further optimize FaceNet model. Finally, the face feature vector is extracted by FaceNet model to realize the correct recognition of mask face. Experiments show that this scheme can improve the accuracy of face recognition with masks to a certain extent, and at the same time improve the traffic speed.