Human Interaction Recognition On Ntu Rgb D
Metrics
Accuracy (Cross-Subject)
Accuracy (Cross-View)
Results
Performance results of various models on this benchmark
Model Name | Accuracy (Cross-Subject) | Accuracy (Cross-View) | Paper Title | Repository |
|---|---|---|---|---|
| SkateFormer | 97.1 | 99.3 | SkateFormer: Skeletal-Temporal Transformer for Human Action Recognition | |
| LSTM-IRN'fc1inter+intra | 90.5 | 93.5 | Interaction Relational Network for Mutual Action Recognition | |
| CHASE(CTR-GCN) | 96.5 | 98.8 | CHASE: Learning Convex Hull Adaptive Shift for Skeleton-based Multi-Entity Action Recognition | |
| IGFormer | 93.6 | 96.5 | IGFormer: Interaction Graph Transformer for Skeleton-based Human Interaction Recognition | - |
| SkeleTR | 94.9 | 97.7 | SkeleTR: Towards Skeleton-based Action Recognition in the Wild | - |
0 of 5 row(s) selected.