Panoptic Scene Graph Generation On Psg
평가 지표
R@20
mR@20
평가 결과
이 벤치마크에서 각 모델의 성능 결과
비교 표
| 모델 이름 | R@20 | mR@20 |
|---|---|---|
| learning-to-compose-dynamic-tree-structures | 20.6 | 9.70 |
| panoptic-scene-graph-generation-with | 26.0 | 26.4 |
| hilo-exploiting-high-low-frequency-relations | 34.1 | 23.7 |
| neural-motifs-scene-graph-parsing-with-global | 20.0 | 9.10 |
| panoptic-scene-graph-generation | 28.4 | 16.6 |
| scene-graph-generation-by-iterative-message | 16.5 | 6.52 |
| hilo-exploiting-high-low-frequency-relations | 40.6 | 29.7 |
| panoptic-scene-graph-generation | 18.0 | 14.8 |
| vlprompt-vision-language-prompting-for | 39.4 | 34.7 |