Skip to main content
Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

< Back to Article

Fig 1.

Residual behavior.

More »

Fig 1 Expand

Fig 2.

Inverted residual bottleneck block.

More »

Fig 2 Expand

Fig 3.

Proposed architecture for human action recognition on selected action datasets.

More »

Fig 3 Expand

Fig 4.

Sample classes from the HMDB51 dataset.

More »

Fig 4 Expand

Fig 5.

Sample classes from UCF101 dataset [40].

More »

Fig 5 Expand

Fig 6.

Proposed inverted residual parallel block.

More »

Fig 6 Expand

Fig 7.

Self-attention module for the features learning.

More »

Fig 7 Expand

Fig 8.

Complete architecture of proposed inverted bottleneck residual with self-attention CNN.

More »

Fig 8 Expand

Fig 9.

Detailed, layered architecture of proposed Inverted Bottleneck Residual with Self-Attention (InBRwSA) for action recognition.

More »

Fig 9 Expand

Fig 10.

Proposed architecture testing phase.

More »

Fig 10 Expand

Table 1.

Proposed architecture classification accuracy using the HMDB51 dataset.

More »

Table 1 Expand

Fig 11.

Confusion matrix for SWNN classifier using proposed architecture.

More »

Fig 11 Expand

Table 2.

Proposed architecture classification accuracy using UCF101 Dataset.

More »

Table 2 Expand

Fig 12.

Confusion matrix for SWNN classifier using proposed architecture using UCF101 dataset.

More »

Fig 12 Expand

Fig 13.

Ablation study 1- pre-trained deep learning-based comparison for the selected datasets.

More »

Fig 13 Expand

Fig 14.

Ablation Study 2- Confidence interval-based analysis of proposed architecture on selected datasets.

More »

Fig 14 Expand

Table 3.

Comparison with existing SOTA techniques.

More »

Table 3 Expand

Fig 15.

Proposed architecture prediction results using video sequences.

More »

Fig 15 Expand