Research Article
Vision Transformer and Deep Sequence Learning for Human Activity Recognition in Surveillance Videos
Table 1
Parameters details shown in the formulation of LSTM network.
| Variables/symbol | Description |
| | Input over time t | | Sigmoid activation function | | Weights | | Bias terms | | Input gate | | Forget gate | | Output gate | | Tan h activation function | | Activation for the final classification | | Numbers of classes |
|
|