Masked Face Recognition Using Deep Learning Techniques
Yiyang Ma, Naruemon Wattanapongsakorn · 2023
Since the COVID-19 pandemic, people have been wearing masks in public, which poses a huge challenge to facial recognition. This paper introduces a new solution which is a masked face recognition method based on FaceNet and Inception-ResNet which are deep learning techniques. A well-known benchmark dataset (RWMFD) is employed to train the face recognition model. Furthermore, simulating masks and data augmentation are used to further improve the classification precision of the model. Our four experimental cases on the synthetic and the real-mask face datasets show that our method can effectively process face recognition and identification task with and without masks, with very high performance in terms of accuracy, precision, recall and F1-score.