Masked Face Recognition using Deep Learning Model
Manoj Kumar, Rachit Mann · 2021 3rd International Conference on Advances in Computing, Communication Control and Networking (ICAC3N) · 2021
Face Masks have become part of our day-to-day activities. But these create a problem with the existing face recognition techniques. Existing face recognition techniques vary from simple ML techniques such as SVM, PCA, etc. to state-of-the-art models such as ResNet, VGG, etc. In this paper, we are studying the effects of masked faces on the performance of face recognition techniques. Face recognition itself is divided into face verification and face identification. This study is performed for the face identification task using different deep learning models. In particular, we are studying the pre-trained deep learning models that are trained using transfer learning for face identification tasks. In this study, a custom dataset is used consisting of 65 subjects. The custom dataset is a part of the VGGFace2 dataset. Subject faces are augmented with masks. The dataset doesn’t consist the faces without masks. In this study, we are using different popular pre-trained models such as VGG16, InceptionV3, etc. We are re-training those pre-trained models on a custom dataset and analyzing the gathered results. Apart from the pre-trained models, a new model has been proposed for the masked face identification task.