Multi-type Digital Recognition Based on TensorFlow

Real-time face mask position recognition system based on MobileNet model

“…To detect faces from digital images, we employ the MTCNN model. Face detection and facial landmark localization in MTCNN are performed jointly through multitask learning in a coarse-to-fine approach, using a three-stage deep CNN to construct a cascaded architecture ( Ma & Wang, 2018 ). The detected faces are then resized to a resolution of 224 × 224 and stored along with the image labels.…”