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.

The general flowchart about behavior recognition.

More »

Fig 1 Expand

Fig 2.

The flowchart about our target detection process.

More »

Fig 2 Expand

Fig 3.

The feature extraction process.

More »

Fig 3 Expand

Fig 4.

The ROI extraction process.

More »

Fig 4 Expand

Fig 5.

The bounding-box regression and category prediction process.

More »

Fig 5 Expand

Fig 6.

The image transformation process.

More »

Fig 6 Expand

Fig 7.

The process of fragment and stochastic sample about video.

More »

Fig 7 Expand

Fig 8.

The left side is inflated-3-dimensional network, and the right side is the inception structure.

More »

Fig 8 Expand

Fig 9.

The influence of different values of α on UCF-101 about accuracy.

More »

Fig 9 Expand

Fig 10.

The influence of different values of α on HMDB-51 about accuracy.

More »

Fig 10 Expand

Table 1.

The effect of different α and different partitions on hmdb-51 about accuracy.

More »

Table 1 Expand

Table 2.

The effect of different α and different partitions on ucf-101 about accuracy.

More »

Table 2 Expand

Table 3.

The effect of different combinations of α and γ on hmdb-51 about accuracy.

More »

Table 3 Expand

Table 4.

The effect of different combinations of α and γ on ucf-101 about accuracy.

More »

Table 4 Expand

Table 5.

The effect of different combinations of input on hmdb-51 about accuracy.

More »

Table 5 Expand

Table 6.

The effect of different combinations of input ucf-101 about accuracy.

More »

Table 6 Expand

Fig 11.

The confusion matrix of the first sectionalization in UCF-101 under the input warped & original image.

More »

Fig 11 Expand

Fig 12.

The confusion matrix of the first sectionalization in UCF-101 under the input cropped & original image.

More »

Fig 12 Expand

Fig 13.

The confusion matrix of the first sectionalization in UCF-101 under the input cropped image.

More »

Fig 13 Expand

Fig 14.

The confusion matrix of the first sectionalization in HMDB-51 under the input warped & original image.

More »

Fig 14 Expand

Fig 15.

The confusion matrix of the first sectionalization in HMDB-51 under the input cropped & original image.

More »

Fig 15 Expand

Fig 16.

The confusion matrix of the first sectionalization in HMDB-51 under the input cropped image.

More »

Fig 16 Expand

Fig 17.

Some typical class accuracy display under different inputs on HMDB-51.

More »

Fig 17 Expand

Fig 18.

Some typical class accuracy display under different inputs on UCF-101.

More »

Fig 18 Expand

Table 7.

Accuracy comparison of different methods in UCF-101 and HMDB-51.

More »

Table 7 Expand

Table 8.

Accuracy comparison of different methods in kinetics.

More »

Table 8 Expand