Skeleton Based Action Recognition On Cad 120
Metrics
Accuracy
Results
Performance results of various models on this benchmark
Model Name | Accuracy | Paper Title | Repository |
|---|---|---|---|
| All Features (w ground truth) | 89.3% | Learning Spatio-Temporal Structure from RGB-D Videos for Human Activity Detection and Anticipation | - |
| S-RNN (5-shot) | 85.4% | Structural-RNN: Deep Learning on Spatio-Temporal Graphs | - |
| P-LSTM (5-shot) | 68.1% | NTU RGB+D: A Large Scale Dataset for 3D Human Activity Analysis | - |
| NGM w/o Edges (5-shot) | 85.0% | Neural Graph Matching Networks for Fewshot 3D Action Recognition | - |
| KGS | 86.0% | Learning Human Activities and Object Affordances from RGB-D Videos | - |
| PointNet (5-shot) | 69.1% | PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation | - |
| NGM (5-shot) | 91.1% | Neural Graph Matching Networks for Fewshot 3D Action Recognition | - |
| Our DP seg. + moves + heuristic seg. | 70.3% | Learning Spatio-Temporal Structure from RGB-D Videos for Human Activity Detection and Anticipation | - |
0 of 8 row(s) selected.