Vision Transformer and Deep Sequence Learning for Human Activity Recognition in Surveillance Videos

Figure 4

Confusion matrix of the proposed model. (a) UCF50 and (b) HMDB51 dataset.

(a)

(b)