Object Detection on MS-COCO 2017 (val) (Base/Novel Split AP50)
59.4Base Avg AP50REACT
Evaluation Results
| Method | Links | ||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| REACTEvaluation Setting=Zero-Shot, Backbone=ResNet-50, Region Proposals=GT, Pipeline=RegionCLIP2023.01 | 59.4 | — | — | — | — | — | — | — | — | 58.9 | 59.3 | — | — | — | |
| Base-Onlysetting=gFSIS2025.05 | 59.25 | — | — | — | — | — | — | — | — | — | — | — | 39.86 | — | |
| CLIPEvaluation Setting=Zero-Shot, Backbone=ResNet-50, Region Proposals=GT, Pipeline=RegionCLIP2023.01 | 58.2 | — | — | — | — | — | — | — | — | 58.6 | 58.3 | — | — | — | |
| REACTEvaluation Setting=OVD, Backbone=ResNet-50, Pipeline=RegionCLIP2023.01 | 55.1 | — | — | — | — | — | — | — | — | 20.6 | 46.1 | — | — | — | |
| Decoupling ClassifierShots=30, setting=gFSIS2025.05 | 55.05 | — | — | — | — | — | — | — | — | 40.31 | 51.37 | 32.92 | 36.45 | 22.3 | |
| Decoupling ClassifierShots=10, setting=gFSIS2025.05 | 54.85 | — | — | — | — | — | — | — | — | 34.55 | 49.77 | 31.77 | 36.14 | 18.67 | |
| Decoupling ClassifierShots=5, setting=gFSIS2025.05 | 53.65 | — | — | — | — | — | — | — | — | 30.05 | 47.75 | 30.48 | 35.3 | 16.02 | |
| Decoupling ClassifierShots=3, setting=gFSIS2025.05 | 53.3 | — | — | — | — | — | — | — | — | 24.95 | 46.21 | 29.59 | 35.07 | 13.15 | |
| Decoupling ClassifierShots=2, setting=gFSIS2025.05 | 52.82 | — | — | — | — | — | — | — | — | 20.49 | 44.74 | 28.63 | 34.67 | 10.52 | |
| CLIPEvaluation Setting=OVD, Backbone=ResNet-50, Pipeline=RegionCLIP2023.01 | 52.8 | — | — | — | — | — | — | — | — | 14.2 | 42.7 | — | — | — | |
| Decoupling ClassifierShots=1, setting=gFSIS2025.05 | 52.46 | — | — | — | — | — | — | — | — | 12.79 | 42.55 | 27.35 | 34.35 | 6.34 | |
| Mask-DeFRCN†Shots=30, setting=gFSIS2025.05 | 52.01 | — | — | — | — | — | — | — | — | 36.44 | 48.11 | 31.66 | 35.1 | 21.33 | |
| Mask-DeFRCN†Shots=10, setting=gFSIS2025.05 | 50.48 | — | — | — | — | — | — | — | — | 29.58 | 45.25 | 29.88 | 34.17 | 17.02 | |
| SaFTBackbone=ResNet-1012022.03 | 48.3 | 49.2 | 47.2 | 47.9 | 49 | 27.8 | 27.6 | 21 | 23 | 24.9 | — | — | — | — | |
| Mask-DeFRCN†Shots=5, setting=gFSIS2025.05 | 48.03 | — | — | — | — | — | — | — | — | 24.41 | 42.12 | 27.82 | 32.54 | 13.69 | |
| AIT2022.03 | 47.5 | 50.1 | 47.2 | 45.8 | 46.9 | 26 | 26.4 | 22.3 | 22.6 | 24.3 | — | — | — | — | |
| REACTEvaluation Setting=Zero-Shot, Backbone=ViT-B/16, Region Proposals=GT, Pipeline=RegionCLIP2023.01 | 47.2 | — | — | — | — | — | — | — | — | 43.5 | 46.3 | — | — | — | |
| Mask-DeFRCN†Shots=3, setting=gFSIS2025.05 | 46.83 | — | — | — | — | — | — | — | — | 19.55 | 40.01 | 26.54 | 31.77 | 10.87 | |
| Mask-DeFRCN†Shots=2, setting=gFSIS2025.05 | 45.82 | — | — | — | — | — | — | — | — | 15.79 | 38.31 | 25.42 | 31.06 | 8.52 | |
| Mask-DeFRCN†Shots=1, setting=gFSIS2025.05 | 44.42 | — | — | — | — | — | — | — | — | 9.55 | 35.7 | 23.82 | 30.11 | 4.95 | |
| COAE2022.03 | 40.9 | 42.2 | 40.2 | 39.9 | 41.3 | 23.4 | 23.6 | 20.5 | 20.4 | 22 | — | — | — | — | |
| iMTFAShots=1, setting=gFSIS2025.05 | 40.11 | — | — | — | — | — | — | — | — | 5.89 | 31.55 | 21.67 | 27.81 | 3.23 | |
| CLIPEvaluation Setting=Zero-Shot, Backbone=ViT-B/16, Region Proposals=GT, Pipeline=RegionCLIP2023.01 | 38.5 | — | — | — | — | — | — | — | — | 40.2 | 38.9 | — | — | — | |
| iMTFAShots=5, setting=gFSIS2025.05 | 33.69 | — | — | — | — | — | — | — | — | 11.15 | 28.06 | 19.62 | 24.13 | 6.07 | |
| iMTFAShots=10, setting=gFSIS2025.05 | 32.41 | — | — | — | — | — | — | — | — | 12.72 | 27.49 | 19.26 | 23.36 | 6.97 | |
| REACTEvaluation Setting=Zero-Shot, Backbone=ResNet-50, Region Proposals=RPN, Pipeline=RegionCLIP2023.01 | 25.4 | — | — | — | — | — | — | — | — | 31.6 | 27 | — | — | — | |
| CLIPEvaluation Setting=Zero-Shot, Backbone=ResNet-50, Region Proposals=RPN, Pipeline=RegionCLIP2023.01 | 24 | — | — | — | — | — | — | — | — | 29.7 | 25.5 | — | — | — |