Skip to main content
. 2022 Sep 26;10:955332. doi: 10.3389/fpubh.2022.955332

Table 5.

Summary of works carried out using Hybrid FMD.

Reference Methodology Dataset Performance
Bhattacharya (55) HybridFace maskNet 62% accuracy
Loey et al. (56) Resnet50
SVM
Ensemble algorithm
Decision tree
Real-World Masked Face Dataset (RMFD)
Simulated Masked Face Dataset (SMFD)
Labeled Faces in the Wild (LFW)
SVM 99.64% had accuracy on RMFD
99.49% accuracy on SMFD and 100% accuracy on LFW
Aydemir et al. (57) Pre-trained ResNet101 and DenseNet201 95.95, 97.49, and 100.0% accuracies for the three cases.
Ristea and Ionescu (58) ResNet
SVM
74.6% accuracy
Nieto-Rodríguez et al. (59) Viola-Jones face detection technique BAO database
Wang et al. (41) RCNN and InceptionV2 WMD 97.32% accuracy for basic scenarios
91.13% accuracy for more difficult scenarios