Table 1.

Summarized information of the state-of-the-art work

Contributors	Purpose	Techniques	Results
Asghar et al. [2]	Face mask detection.	Depth-wise separable convolution layers based on MobileNet is applied instead of 2D convolution layers. It is applied on AIZOO face mask dataset.	Acc. = 93.14, Pre. = 0.92, Recall = 0.92, F-score = 0.92
Teboulbi et al. [21]	Face mask detection.	Pretrained models such as the MobileNet, ResNet Classifier, and VGG are used on Kaggle face mask dataset.	Acc. = 97.14, Pre. = 0.982, Recall = 0.973, F-score = 0.96
Goyal et al. [7]	Face mask detection.	Convolution Neural Network layers (CNN) are used as its backbone architecture to create different layers. Kaggle face mask dataset is used.	Acc. = 98.0, Pre. = 0.98., Recall = 0.97, F-score = 0.98
Kumar et al. [11]	Face mask detection.	Raspberry Pi circuit with CNN model are used on Kaggle face mask dataset is used.	Acc. = 97.01, Pre. = 0.98, Recall = 0.98
Sethi et al. [15]	Face mask detection.	Three popular baseline models viz., ResNet50, AlexNet and MobileNet are on Kaggle face mask dataset	Acc. = 98.20, Pre. = 0.98, Recall = 0.98
Das et al. [6]	Face mask detection.	Prepared a CNN model with 200 filters and reshaping of the input images are done.	Acc. = 95.77
Taneja et al. [20]	Face mask detection.	Lightweight MobileNetV2 architecture is used on Kaggle face mask dataset.	Acc. = 99.02
Hussain et al. [9]	Face mask detection.	Apply deep convolution neural network (DCNN) and MobileNetV2-based transfer learning models.	Acc. =98.00, Pre. = 0.98, Recall = 0.98