Skeleton Based Action Recognition On Varying
Metrics
Accuracy (AV I)
Accuracy (AV II)
Accuracy (CS)
Accuracy (CV I)
Accuracy (CV II)
Results
Performance results of various models on this benchmark
Model Name | Accuracy (AV I) | Accuracy (AV II) | Accuracy (CS) | Accuracy (CV I) | Accuracy (CV II) | Paper Title | Repository |
|---|---|---|---|---|---|---|---|
| SK-CNN | 43% | 77% | 59% | 26% | 68% | Enhanced skeleton visualization for view invariant human action recognition | - |
| TCN | 43% | 64% | 56% | 16% | 43% | Temporal Convolutional Networks for Action Segmentation and Detection | - |
| P-LSTM | 33% | 50% | 60% | 13% | 33% | NTU RGB+D: A Large Scale Dataset for 3D Human Activity Analysis | - |
| ST-GCN | 53% | 43% | 71% | 25% | 56% | Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognition | - |
| Res-TCN | 48% | 68% | 63% | 14% | 48% | Interpretable 3D Human Action Analysis with Temporal Convolutional Networks | - |
| LSTM | 31% | 68% | 56% | 16% | 31% | NTU RGB+D: A Large Scale Dataset for 3D Human Activity Analysis | - |
| VS-CNN | 57% | 75% | 76% | 29% | 71% | A Large-scale Varying-view RGB-D Action Dataset for Arbitrary-view Human Action Recognition | - |
0 of 7 row(s) selected.