Semantic Segmentation on COCO-20i 1-shot
49.5Fold 0 ScoreDCAMA
Evaluation results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| DCAMAImage encoder=Swin-B, Learnable params=47.7M2024.02 | 49.5 | 52.7 | 52.8 | 48.7 | 50.9 | — | — | |
| VRP-SAMImage encoder=ResNet-50, Learnable params=1.6M2024.02 | 48.1 | 55.8 | 60 | 51.6 | 53.9 | — | — | |
| HDMNetImage encoder=ResNet-50, Learnable params=4.2M2024.02 | 43.8 | 55.3 | 51.6 | 49.4 | 50 | — | — | |
| VRP-SAMImage encoder=VGG-16, Learnable params=1.6M2024.02 | 43.6 | 51.7 | 50 | 46.5 | 48 | — | — | |
| BAMBackbone network=ResNet50, # learnable params=26.7M2022.11 | 43.4 | 50.6 | 47.5 | 43.4 | 46.2 | — | — | |
| ProtoFormerBackbone network=ResNet101, # learnable params=0.6M2022.11 | 42.9 | 50.7 | 48.4 | 45.8 | 47 | 70 | — | |
| ProtoFormerBackbone network=ResNet50, # learnable params=0.6M2022.11 | 42.4 | 48.5 | 46.3 | 45.5 | 45.7 | 69.6 | — | |
| DPCNBackbone network=ResNet502022.11 | 42 | 47 | 43.3 | 39.7 | 43 | 63.2 | — | |
| DPCNImage encoder=ResNet-502024.02 | 42 | 47 | 43.3 | 39.7 | 43 | — | — | |
| DCAMABackbone network=ResNet50, # learnable params=47.7M2022.11 | 41.9 | 45.1 | 44.4 | 41.7 | 43.3 | 69.5 | — | |
| DCAMABackbone network=ResNet101, # learnable params=47.7M2022.11 | 41.5 | 46.2 | 45.2 | 41.3 | 43.5 | 69.9 | — | |
| IPMTBackbone network=ResNet502022.11 | 41.4 | 45.1 | 45.6 | 40 | 43 | — | — | |
| HDMNetImage encoder=VGG-16, Learnable params=4.2M2024.02 | 40.7 | 50.6 | 48.2 | 44 | 45.9 | — | — | |
| IPMTBackbone network=ResNet1012022.11 | 40.5 | 45.7 | 44.8 | 39.3 | 42.6 | — | — | |
| BAMImage encoder=ResNet-50, Learnable params=4.9M2024.02 | 39.4 | 49.9 | 46.2 | 45.2 | 45.2 | — | — | |
| SSPBackbone network=ResNet101, # learnable params=27.7M2022.11 | 39.1 | 45.1 | 42.7 | 41.2 | 42 | — | — | |
| VATBackbone network=ResNet50, # learnable params=3.2M2022.11 | 39 | 43.8 | 42.6 | 39.7 | 41.3 | 68.8 | — | |
| VATImage encoder=ResNet-50, Learnable params=3.2M2024.02 | 39 | 43.8 | 42.6 | 39.7 | 41.3 | — | — | |
| CyCTRBackbone network=ResNet50, # learnable params=15.4M2022.11 | 38.9 | 43 | 39.6 | 39.8 | 40.3 | — | — | |
| CyCTRImage encoder=ResNet-50, Learnable params=15.4M2024.02 | 38.9 | 43 | 39.6 | 39.8 | 40.3 | — | — | |
| NTRENetBackbone network=ResNet101, # learnable params=19.9M2022.11 | 38.3 | 40.4 | 39.5 | 38.1 | 39.1 | 67.5 | — | |
| HSNetBackbone network=ResNet101, # learnable params=2.5M2022.11 | 37.2 | 44.1 | 42.4 | 41.3 | 41.2 | 69.1 | — | |
| NTRENetBackbone network=ResNet50, # learnable params=19.9M2022.11 | 36.8 | 42.6 | 39.9 | 37.9 | 39.3 | 68.5 | — | |
| PFENetBackbone network=ResNet101, # learnable params=10.3M2022.11 | 36.8 | 41.8 | 38.7 | 36.7 | 38.5 | 63 | — | |
| NTRENetImage encoder=ResNet-50, Learnable params=19.9M2024.02 | 36.8 | 42.6 | 39.9 | 37.9 | 39.3 | — | — | |
| PFENetBackbone network=ResNet50, # learnable params=10.3M2022.11 | 36.5 | 38.6 | 34.5 | 33.8 | 35.8 | — | — | |
| PFENetImage encoder=ResNet-50, Learnable params=10.4M2024.02 | 36.5 | 38.6 | 34.5 | 33.8 | 35.8 | — | — | |
| SCLBackbone network=ResNet1012022.11 | 36.4 | 38.6 | 37.5 | 35.4 | 37 | — | — | |
| BAMImage encoder=VGG-16, Learnable params=1.9M2024.02 | 36.4 | 47.1 | 43.3 | 41.7 | 42.1 | — | — | |
| HSNetBackbone network=ResNet50, # learnable params=2.5M2022.11 | 36.3 | 43.1 | 38.7 | 38.7 | 39.2 | 68.2 | — | |
| HSNetImage encoder=ResNet-50, Learnable params=2.6M2024.02 | 36.3 | 43.1 | 38.7 | 38.7 | 39.2 | — | — | |
| SAGNNBackbone network=ResNet1012022.11 | 36.1 | 41 | 38.2 | 33.5 | 37.2 | — | — | |
| SSPBackbone network=ResNet50, # learnable params=8.7M2022.11 | 35.5 | 39.6 | 37.9 | 36.7 | 37.4 | — | — | |
| SSPImage encoder=ResNet-50, Learnable params=8.7M2024.02 | 35.5 | 39.6 | 37.9 | 36.7 | 37.4 | — | — | |
| PFENetImage encoder=VGG-16, Learnable params=10.4M2024.02 | 35.4 | 38.1 | 36.8 | 34.7 | 36.3 | — | — | |
| CWTBackbone network=ResNet502022.11 | 32.2 | 36 | 31.6 | 31.6 | 32.9 | — | — | |
| RePRIBackbone network=ResNet502022.11 | 32 | 38.7 | 32.7 | 33.1 | 34.1 | — | — | |
| CWTBackbone network=ResNet1012022.11 | 30.3 | 36.6 | 30.5 | 32.2 | 32.4 | — | — | |
| PMMBackbone network=ResNet502022.11 | 29.3 | 34.8 | 27.1 | 27.3 | 29.6 | — | — | |
| PPNetBackbone network=ResNet50, # learnable params=31.5M2022.11 | 28.1 | 30.8 | 29.5 | 27.7 | 29 | — | — | |
| FWBBackbone network=ResNet101, # learnable params=43.0M2022.11 | 17 | 18 | 21 | 28.9 | 21.2 | — | — | |
| Matchermode=one-shot, encoder=SAM2024.03 | — | — | — | — | — | — | 25.1 | |
| P2SAMmode=one-shot, encoder=SAM2024.03 | — | — | — | — | — | — | 26 | |
| PerSAMmode=one-shot, encoder=SAM2024.03 | — | — | — | — | — | — | 23 |