Image Retrieval on RSICD (test)
11.63R@1Ours
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| OursBackbone=CLIP(ViT-B-32), Params=0.16M2023.08 | 11.63 | 33.92 | 50.73 | |
| Full fine-tuningBackbone=CLIP(ViT-B-32), Params=151M2023.08 | 11.55 | 33.14 | 49.83 | |
| UniAdapterBackbone=CLIP(ViT-B-32), Params=0.55M2023.08 | 9.61 | 30.06 | 47.16 | |
| Cross-Modal AdapterBackbone=CLIP(ViT-B-32), Params=0.16M2023.08 | 9.57 | 30.74 | 48.36 | |
| Single LanguageBackbone=CLIP(ViT-B-32), Params=151M2023.08 | 9.14 | 28.96 | 44.59 | |
| AdaptFormerBackbone=CLIP(ViT-B-32), Params=0.17M2023.08 | 9.09 | 29.89 | 46.81 | |
| AdapterBackbone=CLIP(ViT-B-32), Params=0.17M2023.08 | 8.43 | 26.02 | 43.33 | |
| Linear probeBackbone=CLIP(ViT-B-32), Params=1.05M2023.08 | 7.81 | 25.89 | 42.47 | |
| CLIP-AdapterBackbone=CLIP(ViT-B-32), Params=0.52M2023.08 | 7.67 | 24.87 | 39.73 | |
| ConvpassBackbone=CLIP(ViT-B-32), Params=0.14M2023.08 | 7.03 | 23.08 | 39.15 | |
| VPTBackbone=CLIP(ViT-B-32), Params=0.46M2023.08 | 6.94 | 25.26 | 40.73 | |
| VL-PromptBackbone=CLIP(ViT-B-32), Params=0.47M2023.08 | 6.69 | 22.62 | 36.16 | |
| CABIRBackbone=ResNet152, BERT+biGRU, Params=Not specified2023.08 | 5.42 | 20.77 | 33.58 | |
| GaLR w/o MRBackbone=ResNet18, biGRU, Params=46.89M2023.08 | 5.11 | 19.57 | 31.92 | |
| AMFMN-softBackbone=ResNet18, biGRU, Params=35.94M2023.08 | 5.05 | 19.74 | 31.04 | |
| MCRNBackbone=ResNet18, biGRU, Params=52.35M2023.08 | 5.03 | 19.38 | 32.99 | |
| Zero-shot CLIPBackbone=CLIP(ViT-B-32), Params=0.00M2023.08 | 5.01 | 15.75 | 24.21 | |
| MTFNBackbone=Faster R-CNN, biGRU, Params=77.90M2023.08 | 4.9 | 17.17 | 29.49 | |
| AMFMN-fusionBackbone=ResNet18, biGRU, Params=35.94M2023.08 | 4.9 | 18.28 | 31.44 | |
| GaLR with MRBackbone=ResNet18, biGRU, Params=46.89M2023.08 | 4.69 | 19.48 | 32.13 | |
| LW-MCR-dBackbone=SqueezeNet, -, Params=1.65M2023.08 | 4.66 | 17.51 | 30.02 | |
| CoOpBackbone=CLIP(ViT-B-32), Params=0.10M2023.08 | 4.31 | 17.89 | 31.73 | |
| LW-MCR-uBackbone=SqueezeNet, -, Params=1.65M2023.08 | 4.3 | 18.85 | 32.34 | |
| CAMP-tripletBackbone=Faster R-CNN, biGRU, Params=27.03M2023.08 | 4.15 | 15.23 | 27.81 | |
| AMFMN-simBackbone=ResNet18, biGRU, Params=35.94M2023.08 | 4.08 | 17 | 30.6 | |
| LW-MCR-bBackbone=SqueezeNet, -, Params=1.65M2023.08 | 4.02 | 16.47 | 28.23 | |
| SCAN t-iBackbone=Faster R-CNN, biGRU, Params=13.68M2023.08 | 3.91 | 16.2 | 26.49 | |
| SCAN i-tBackbone=Faster R-CNN, biGRU, Params=13.68M2023.08 | 3.71 | 16.4 | 26.73 | |
| VSE++Backbone=ResNet18, GRU, Params=15.78M2023.08 | 2.82 | 11.32 | 18.1 | |
| CAMP-bceBackbone=Faster R-CNN, biGRU, Params=27.03M2023.08 | 2.72 | 12.76 | 22.89 |