Fig 1.
(a) Mask or with out mask. (b) Cover Large parts of their Face. (c) Walking.
Fig 2.
Face detection work.
Fig 3.
Gait recognition work.
Fig 4.
Our proposed work.
Fig 5.
Human face dataset description.
Fig 6.
Gait recognition (CASIA-A) dataset [39].
Table 1.
Hyberparameters of CNN architectures.
Fig 7.
Machine and deep learning results.
(a) Machine learning results. (b) Deep learning results.
Fig 8.
(a) AlexNet Face. (b) AlexNet Gait.
Table 2.
Performance assessments of different detection deep learning architectures for gait (CASIA-A) dataset.
Fig 9.
(a) VGG16 Face. (b) VGG16 Gait.
Fig 10.
(a) VGG19 Face. (b) VGG19 Gait.
Fig 11.
(a) CNN Face. (b) CNN Gait.
Table 3.
Performance assessments of different detection deep learning architectures for human masked face dataset.
Table 4.
Comparison between our approach and previous research results based on Human face detection.
Table 5.
Comparison between our approach and previous research results based on Human gait detection.
Fig 12.
VGG16 and 19 architectures.
Table 6.
Results of scratch and transfer learning.
Table 7.
Performance assessments of different detection deep learning architectures for human masked face and gait datasets.
Table 8.
Performance assessments of different detection machine learning algorithms using deep learning features for human masked face and gait datasets.
Table 9.
Execution time performances of deep learning architectures.
Table 10.
Execution time performances of machine learning classifiers.