Multimodal Activity Recognition On Ev Action
Metrics
Accuracy
Results
Performance results of various models on this benchmark
| Paper Title | ||
|---|---|---|
| TCN (Skeleton Kinect) | 80.1 | Interpretable 3D Human Action Analysis with Temporal Convolutional Networks |
| ST-GCN (Skeleton Kinect) | 79.6 | Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognition |
| TSN (RGB) | 73.6 | Temporal Segment Networks: Towards Good Practices for Deep Action Recognition |
| TCN-RMS (Skeleton Kinect+EMG) | 67.4 | EV-Action: Electromyography-Vision Multi-Modal Action Dataset |
| TCN-FFT (Skeleton Vicon+EMG) | 64.4 | EV-Action: Electromyography-Vision Multi-Modal Action Dataset |
| TCN (Skeleton Vicon) | 64.1 | Interpretable 3D Human Action Analysis with Temporal Convolutional Networks |
| ST-GCN (Skeleton Vicon) | 50.7 | Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognition |
| LSTM-FFT (EMG) | 44.1 | EV-Action: Electromyography-Vision Multi-Modal Action Dataset |
| WHDMM (Depth) | 40.2 | Action recognition from depth maps using deep convolutional neural networks |
0 of 9 row(s) selected.