Scene Graph Anticipation on VSGR (test)
30.2R@10HyperGLM
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| HyperGLMVideo input fraction (F)=0.9, Pre-trained detector=Faster R-CNN2024.11 | 30.2 | 18.1 | 41.1 | 23.5 | 59.3 | 43.4 | |
| HyperGraphVideo input fraction (F)=0.9, Pre-trained detector=Faster R-CNN2024.11 | 28.4 | 17.5 | 39.3 | 22.4 | 57.6 | 41.5 | |
| SceneSayerSDEVideo input fraction (F)=0.9, Pre-trained detector=Faster R-CNN2024.11 | 27.5 | 16 | 38.5 | 22 | 58.2 | 40 | |
| SceneSayerODEVideo input fraction (F)=0.9, Pre-trained detector=Faster R-CNN2024.11 | 26.5 | 14 | 37.5 | 20 | 55.5 | 38 | |
| DSGDetr++Video input fraction (F)=0.9, Pre-trained detector=Faster R-CNN2024.11 | 19.5 | 10 | 31 | 16 | 50 | 32.5 | |
| STTran++Video input fraction (F)=0.9, Pre-trained detector=Faster R-CNN2024.11 | 19 | 9.5 | 30 | 15.5 | 49 | 31 | |
| DSGDetr+Video input fraction (F)=0.9, Pre-trained detector=Faster R-CNN2024.11 | 17 | 6 | 27 | 11.5 | 44.5 | 24.5 | |
| STTran+Video input fraction (F)=0.9, Pre-trained detector=Faster R-CNN2024.11 | 16.5 | 5.5 | 26 | 11 | 43 | 23 |