Skeleton Based Action Recognition On Shrec
Metrics
14 gestures accuracy
28 gestures accuracy
Speed (FPS)
Results
Performance results of various models on this benchmark
Model Name | 14 gestures accuracy | 28 gestures accuracy | Speed (FPS) | Paper Title | Repository |
|---|---|---|---|---|---|
| MFA-Net | 91.3 | 86.6 | 361 | Motion feature augmented network for dynamic hand gesture recognition from skeletal data | - |
| STA-Res-TCN | 93.6 | 90.7 | 161 | Spatial-Temporal Attention Res-TCN for Skeleton-Based Dynamic Hand Gesture Recognition | - |
| TD-GCN | 97.02 | 95.36 | - | Temporal Decoupling Graph Convolutional Network for Skeleton-based Gesture Recognition | - |
| TCN-Summ | 93.57 | 91.43 | - | Domain and View-point Agnostic Hand Action Recognition | - |
| HT-ConvNet | 97.1 | 94.3 | - | Hierarchical Temporal Convolution Network:Towards Privacy-Centric Activity Recognition |
0 of 5 row(s) selected.