Situation Recognition on imSitu (test)
73.41Value AccuracyCAQ w/ RE-VGG
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| CAQ w/ RE-VGGEvaluation Setting=Ground-Truth Verb2022.03 | 73.41 | — | 38.52 | |
| Kernel GraphNetEvaluation Setting=Ground-Truth Verb2022.03 | 72.92 | — | 42.35 | |
| CRF w/ DataAugEvaluation Setting=Ground-Truth Verb2022.03 | 70.44 | — | 34.38 | |
| RNN w/ FusionEvaluation Setting=Ground-Truth Verb2022.03 | 70.27 | — | 35.25 | |
| GraphNetEvaluation Setting=Ground-Truth Verb2022.03 | 69.16 | — | 41.36 | |
| CRFEvaluation Setting=Ground-Truth Verb2022.03 | 65.66 | — | 28.96 | |
| Kernel GraphNetEvaluation Setting=Top-5 Predicted Verbs2022.03 | 55.62 | 68.72 | 30.29 | |
| Ours (Attention Feature Augmentation)2025.01 | 51.1 | 58.88 | 31.56 | |
| CAQ w/ RE-VGGEvaluation Setting=Top-5 Predicted Verbs2022.03 | 50.21 | 65.05 | 28.93 | |
| ClipSitu XTF2025.01 | 47.23 | 58.19 | 29.73 | |
| CRF w/ DataAugEvaluation Setting=Top-5 Predicted Verbs2022.03 | 46.88 | 62.59 | 25.46 | |
| RNN w/ FusionEvaluation Setting=Top-5 Predicted Verbs2022.03 | 46.88 | 63.08 | 26.06 | |
| GraphNetEvaluation Setting=Top-5 Predicted Verbs2022.03 | 45.39 | 61.9 | 29.96 | |
| CRFEvaluation Setting=Top-5 Predicted Verbs2022.03 | 42.76 | 58.88 | 22.55 | |
| CoFormer2025.01 | 35.98 | 44.66 | 22.22 | |
| Kernel GraphNetEvaluation Setting=Top-1 Predicted Verb2022.03 | 35.41 | 43.27 | 19.38 | |
| SituFormer2025.01 | 35.24 | 44.2 | 21.86 | |
| CAQ w/ RE-VGGEvaluation Setting=Top-1 Predicted Verb2022.03 | 30.23 | 38.19 | 18.47 | |
| GraphNetEvaluation Setting=Top-1 Predicted Verb2022.03 | 27.52 | 36.72 | 19.25 | |
| RNN w/ FusionEvaluation Setting=Top-1 Predicted Verb2022.03 | 27.45 | 35.9 | 16.36 | |
| CRF w/ DataAugEvaluation Setting=Top-1 Predicted Verb2022.03 | 26.45 | 34.12 | 15.51 | |
| CRFEvaluation Setting=Top-1 Predicted Verb2022.03 | 24.64 | 32.34 | 14.19 |