Face Detection and Recognition Using Face Mesh and Deep Neural Network

Abstract

Face detection and recognition are emerging and active research areas in computer vision and deep learning. Face detection and recognition have a wide range of applications such as recognizing the people in particular areas namely stores, and banks. Identifying people in a particular database (police database). To control people's entry into restricted areas or grant access to ATMs or computers. In this paper, the proposed model can detect and recognize the face using Face mesh. Due to Face mesh, the model operates in a variety of conditions such as varying illumination and background. The model can also handle non-frontal images of males and females of all ages and races. The Labeled wild face (LWF) dataset images and images captured in real-time are used to train the deep neural network of the model. During testing, if the face landmarks of the test image match with the face landmarks of any of the training images the model gives the name of the person else model outputs as “unknown”.94.23% accuracy are achieved for face recognition by the proposed model.