Situation Recognition On Imsitu
評価指標
Top-1 Verb
Top-1 Verb u0026 Value
Top-5 Verbs
Top-5 Verbs u0026 Value
評価結果
このベンチマークにおける各モデルのパフォーマンス結果
比較表
| モデル名 | Top-1 Verb | Top-1 Verb u0026 Value | Top-5 Verbs | Top-5 Verbs u0026 Value |
|---|---|---|---|---|
| attention-based-context-aware-reasoning-for | 38.19 | 30.23 | 65.05 | 50.21 |
| clipsitu-effectively-leveraging-clip-for | 47.23 | 29.73 | 85.69 | 68.42 |
| collaborative-transformers-for-grounded | 44.66 | 35.98 | 73.31 | 57.76 |
| rethinking-the-two-stage-framework-for | 44.2 | 35.24 | 71.21 | 55.75 |
| grounded-situation-recognition | 39.94 | 31.44 | 67.6 | 51.88 |
| situation-recognition-with-graph-neural | 36.72 | 27.52 | 61.90 | 45.39 |
| commonly-uncommon-semantic-sparsity-in | 34.12 | 26.45 | 62.59 | 46.88 |
| recurrent-models-for-situation-recognition | 35.9 | 27.45 | 63.08 | 46.88 |
| situation-recognition-visual-semantic-role | 32.34 | 24.64 | 58.88 | 42.76 |
| grounded-situation-recognition | 39.36 | 30.09 | 65.51 | 50.16 |
| dynamic-scene-understanding-from-vision | 58.88 | - | - | - |
| grounded-situation-recognition-with | 40.63 | 32.15 | 69.81 | 54.13 |
| mixture-kernel-graph-attention-network-for | 43.27 | 35.41 | 68.72 | 55.62 |