Semantic Segmentation on PASCAL 1-shot 5i
78.3mIoU (fold1)VRP-SAM
Evaluation results
| Method | Links | |||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| VRP-SAMImage encoder=ResNet-50, Learnable params=1.6M2024.02 | 78.3 | 73.9 | 70.6 | 65 | — | — | — | — | — | — | — | — | — | — | 71.9 | |
| HDMNetImage encoder=ResNet-50, Learnable params=4.2M2024.02 | 75.4 | 71 | 68.9 | 62.1 | — | — | — | — | — | — | — | — | — | — | 69.4 | |
| VRP-SAMImage encoder=VGG-16, Learnable params=1.6M2024.02 | 74.7 | 70 | 68.3 | 61.9 | — | — | — | — | — | — | — | — | — | — | 68.7 | |
| DCAMAImage encoder=Swin-B, Learnable params=47.7M2024.02 | 73.8 | 72.2 | 64.3 | 67.1 | — | — | — | — | — | — | — | — | — | — | 69.3 | |
| BAMImage encoder=ResNet-50, Learnable params=4.9M2024.02 | 73.6 | 69 | 67.6 | 61.1 | — | — | — | — | — | — | — | — | — | — | 67.8 | |
| MSDNetBackbone=ResNet101, # learnable params=1.55M2024.09 | 72.8 | 67.6 | 58.2 | 60 | 64.7 | 77.3 | — | — | — | — | — | — | — | — | — | |
| NTRENetBackbone=ResNet50, # learnable params=15.4M2024.09 | 72.3 | 65.4 | 59.4 | 59.8 | 64.2 | 77 | — | — | — | — | — | — | — | — | — | |
| HSNetBackbone=ResNet101, # learnable params=2.5M2024.09 | 72.3 | 67.3 | 62 | 63.1 | 66.2 | 77.6 | — | — | — | — | — | — | — | — | — | |
| NTRENetImage encoder=ResNet-50, Learnable params=19.9M2024.02 | 72.3 | 65.4 | 59.4 | 59.8 | — | — | — | — | — | — | — | — | — | — | 64.2 | |
| VATImage encoder=ResNet-50, Learnable params=3.2M2024.02 | 72 | 67.6 | 62.3 | 60.1 | — | — | — | — | — | — | — | — | — | — | 65.5 | |
| MSDNetBackbone=ResNet50, # learnable params=1.55M2024.09 | 71.9 | 66.3 | 57.2 | 62 | 64.3 | 77.1 | — | — | — | — | — | — | — | — | — | |
| DPCNImage encoder=ResNet-502024.02 | 71.6 | 65.7 | 69.1 | 60.6 | — | — | — | — | — | — | — | — | — | — | 66.7 | |
| HDMNetImage encoder=VGG-16, Learnable params=4.2M2024.02 | 71.4 | 64.8 | 67.7 | 56.4 | — | — | — | — | — | — | — | — | — | — | 65.1 | |
| CyCTRImage encoder=ResNet-50, Learnable params=15.4M2024.02 | 71 | 65.7 | 59.5 | 59.7 | — | — | — | — | — | — | — | — | — | — | 64 | |
| ABCNetBackbone=ResNet50, # learnable params=19.9M2024.09 | 70.8 | 62.5 | 57.2 | 58.1 | 62.2 | 74.1 | — | — | — | — | — | — | — | — | — | |
| BAMImage encoder=VGG-16, Learnable params=1.9M2024.02 | 70.8 | 63.2 | 66.1 | 57.5 | — | — | — | — | — | — | — | — | — | — | 64.4 | |
| HSNetBackbone=ResNet50, # learnable params=2.5M2024.09 | 70.7 | 64.3 | 60.3 | 60.5 | 64 | 76.7 | — | — | — | — | — | — | — | — | — | |
| HSNetImage encoder=ResNet-50, Learnable params=2.6M2024.02 | 70.7 | 64.3 | 60.3 | 60.5 | — | — | — | — | — | — | — | — | — | — | 64 | |
| PFENetBackbone=ResNet50, # learnable params=10.3M2024.09 | 69.5 | 61.7 | 55.4 | 56.3 | 60.8 | 73.3 | — | — | — | — | — | — | — | — | — | |
| PFENetImage encoder=ResNet-50, Learnable params=10.4M2024.02 | 69.5 | 61.7 | 55.4 | 56.3 | — | — | — | — | — | — | — | — | — | — | 60.8 | |
| PFENetImage encoder=VGG-16, Learnable params=10.4M2024.02 | 68.2 | 56.9 | 54.5 | 52.4 | — | — | — | — | — | — | — | — | — | — | 58 | |
| SSPImage encoder=ResNet-50, Learnable params=8.7M2024.02 | 67.8 | 60.5 | 66.4 | 51 | — | — | — | — | — | — | — | — | — | — | 61.4 | |
| PMMsBackbone=ResNet-50, Shots=1-shot, Ways=1-way2020.08 | 67.54 | 51.98 | 51.54 | 49.81 | 55.22 | — | — | — | — | — | — | — | — | — | — | |
| RPMMsBackbone=ResNet-50, Shots=1-shot, Ways=1-way2020.08 | 66.91 | 55.15 | 52.61 | 50.68 | 56.34 | — | — | — | — | — | — | — | — | — | — | |
| PGNetBackbone=ResNet50, # learnable params=17.2M2024.09 | 66.9 | 56 | 50.6 | 50.4 | 56 | 69.9 | — | — | — | — | — | — | — | — | — | |
| RPMMsBackbone=VGG16, Shots=1-shot, Ways=1-way2020.08 | 65.82 | 47.14 | 50.57 | 48.54 | 53.02 | — | — | — | — | — | — | — | — | — | — | |
| CANetBackbone=ResNet-50, Shots=1-shot, Ways=1-way, Test scale=multi-scale2020.08 | 64.97 | 49.56 | 49.83 | 51.49 | 53.96 | — | — | — | — | — | — | — | — | — | — | |
| FWBBackbone=ResNet101, # learnable params=43.0M2024.09 | 64.5 | 51.3 | 56.7 | 52.2 | 56.2 | 70.9 | — | — | — | — | — | — | — | — | — | |
| PFNetBackbone=ResNet-502020.03 | 61.7 | — | 69.5 | 55.4 | — | — | — | — | — | — | — | — | 56.3 | — | 60.8 | |
| PFNetBackbone=ResNet-1012020.03 | 60.5 | — | 69.4 | 54.4 | — | — | — | — | — | — | — | — | 55.9 | — | 60.1 | |
| FWBBackbone=VGG16, Shots=1-shot, Ways=1-way2020.08 | 59.64 | 47.04 | 52.51 | 48.27 | 51.9 | — | — | — | — | — | — | — | — | — | — | |
| SG-OneBackbone=VGG16, Shots=1-shot, Ways=1-way2020.08 | 58.4 | 40.2 | 48.4 | 38.4 | 46.3 | — | — | — | — | — | — | — | — | — | — | |
| PANetBackbone=VGG16, Shots=1-shot, Ways=1-way2020.08 | 58 | 42.3 | 51.1 | 41.2 | 48.1 | — | — | — | — | — | — | — | — | — | — | |
| PANetBackbone=ResNet50, # learnable params=23.5M2024.09 | 57.5 | 44 | 50.8 | 44 | 49.1 | — | — | — | — | — | — | — | — | — | — | |
| DoG-LSTMBackbone=ResNet-1012020.03 | 57 | — | 67.2 | 56.1 | — | — | — | — | — | — | — | — | 54.3 | — | 58.7 | |
| PFNetBackbone=VGG-162020.03 | 56.9 | — | 68.2 | 54.4 | — | — | — | — | — | — | — | — | 52.4 | — | 58 | |
| DoG-LSTMBackbone=VGG-162020.03 | 56.2 | — | 66 | 56.1 | — | — | — | — | — | — | — | — | 53.8 | — | 58 | |
| PGNetBackbone=ResNet-502020.03 | 56 | — | 66.9 | 50.6 | — | — | — | — | — | — | — | — | 50.4 | — | 56 | |
| OSLSMBackbone=VGG16, Shots=1-shot, Ways=1-way2020.08 | 55.3 | 33.6 | 40.9 | 33.5 | 40.8 | — | — | — | — | — | — | — | — | — | — | |
| RPMMBackbone=ResNet-502020.03 | 55.2 | — | 66.9 | 52.6 | — | — | — | — | — | — | — | — | 50.7 | — | 56.3 | |
| DANBackbone=ResNet-1012020.03 | 54.7 | — | 68.6 | 57.8 | — | — | — | — | — | — | — | — | 51.6 | — | 58.2 | |
| LTMBackbone=ResNet-502020.03 | 52.8 | — | 69.6 | 53.2 | — | — | — | — | — | — | — | — | 52.3 | — | 57 | |
| CANetBackbone=ResNet-502020.03 | 52.5 | — | 65.9 | 51.3 | — | — | — | — | — | — | — | — | 51.9 | — | 55.4 | |
| FWBBackbone=ResNet-1012020.03 | 51.3 | — | 64.5 | 56.7 | — | — | — | — | — | — | — | — | 52.2 | — | 56.2 | |
| co-FCNBackbone=VGG16, Shots=1-shot, Ways=1-way2020.08 | 50.6 | 36.7 | 44.9 | 32.4 | 41.1 | — | — | — | — | — | — | — | — | — | — | |
| PPNetBackbone=ResNet-50, report_with_additional_unlabeled_data=false2020.03 | 47.8 | — | 58.8 | 53.8 | — | — | — | — | — | — | — | — | 45.6 | — | 51.5 | |
| RPMMBackbone=VGG-162020.03 | 47.1 | — | 65.8 | 50.6 | — | — | — | — | — | — | — | — | 48.5 | — | 53 | |
| FWBBackbone=VGG-162020.03 | 47 | — | 59.6 | 52.6 | — | — | — | — | — | — | — | — | 48.3 | — | 51.9 | |
| OSAdvBackbone=VGG-162020.03 | 46.9 | — | 59.2 | 49.3 | — | — | — | — | — | — | — | — | 43.4 | — | 49.7 | |
| ARNetBackbone=VGG-162020.03 | 42.6 | — | 59.5 | 50.2 | — | — | — | — | — | — | — | — | 40.2 | — | 48.1 | |
| PANetBackbone=VGG-162020.03 | 42.3 | — | 58 | 51.1 | — | — | — | — | — | — | — | — | 41.2 | — | 48.1 | |
| Meta-SegBackbone=VGG-162020.03 | 42.2 | — | 59.6 | 48.1 | — | — | — | — | — | — | — | — | 44.4 | — | 48.6 | |
| AMPBackbone=VGG-162020.03 | 41.9 | — | 50.2 | 46.7 | — | — | — | — | — | — | — | — | 34.7 | — | 43.4 | |
| SG-OneBackbone=VGG-162020.03 | 40.2 | — | 58.4 | 48.4 | — | — | — | — | — | — | — | — | 38.4 | — | 46.3 | |
| MDLBackbone=VGG-162020.03 | 39.7 | — | 58.3 | 46.7 | — | — | — | — | — | — | — | — | 36.3 | — | 45.3 | |
| co-FCNBackbone=VGG-162020.03 | 36.7 | — | 50.6 | 44.9 | — | — | — | — | — | — | — | — | 32.4 | — | 41.1 | |
| OSLSMBackbone=VGG-162020.03 | 33.6 | — | 55.3 | 40.9 | — | — | — | — | — | — | — | — | 33.5 | — | 40.8 | |
| A-MCGShot=1-shot2019.08 | — | — | — | — | — | — | — | — | — | — | — | 61.2 | — | — | — | |
| A-MCGShot=5-shot2019.08 | — | — | — | — | — | — | — | — | — | — | — | 62.2 | — | — | — | |
| CANetShots=1-shot2019.03 | — | — | — | — | — | — | 71 | 76.7 | 54 | 67.2 | 66.2 | — | — | — | — | |
| CANetShots=5-shot2019.03 | — | — | — | — | — | — | 74.2 | 80.3 | 57 | 66.8 | 69.6 | — | — | — | — | |
| CLIP-Deconv (PC+)t=0.2, vis. backb.=ViT (CLIP), Trained on=PhraseCut+2021.12 | — | — | — | — | 48 | — | — | — | — | — | — | 65.8 | — | 68 | — | |
| CLIPSeg (PC)t=0.3, vis. backb.=ViT (CLIP), Trained on=PhraseCut2021.12 | — | — | — | — | 52.3 | — | — | — | — | — | — | 69.5 | — | 72.4 | — | |
| CLIPSeg (PC+)t=0.3, vis. backb.=ViT (CLIP), Trained on=PhraseCut+2021.12 | — | — | — | — | 59.5 | — | — | — | — | — | — | 75 | — | 82.3 | — | |
| co-FCNShots=1-shot2019.03 | — | — | — | — | — | — | — | — | — | — | 60.1 | — | — | — | — | |
| co-FCNShots=5-shot2019.03 | — | — | — | — | — | — | — | — | — | — | 60.2 | — | — | — | — | |
| co-FCNShot=1-shot2019.08 | — | — | — | — | — | — | — | — | — | — | — | 60.1 | — | — | — | |
| co-FCNShot=5-shot2019.08 | — | — | — | — | — | — | — | — | — | — | — | 60.2 | — | — | — | |
| CRNetBackbone=VGG-162020.03 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 55.2 | |
| CRNetBackbone=ResNet-502020.03 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 55.7 | |
| FG-BGShot=1-shot2019.08 | — | — | — | — | — | — | — | — | — | — | — | 55 | — | — | — | |
| Fine-tuningShot=1-shot2019.08 | — | — | — | — | — | — | — | — | — | — | — | 55.1 | — | — | — | |
| Fine-tuningShot=5-shot2019.08 | — | — | — | — | — | — | — | — | — | — | — | 55.6 | — | — | — | |
| HSNetvis. backb.=RN502021.12 | — | — | — | — | 64 | — | — | — | — | — | — | 76.7 | — | — | — | |
| HSNetvis. backb.=RN1012021.12 | — | — | — | — | 66.2 | — | — | — | — | — | — | 77.6 | — | — | — | |
| MiningFSSFPS=27.8, Params=8.7 M2021.03 | — | — | — | — | 62.1 | — | — | — | — | — | — | — | — | — | — | |
| OSLSMShots=1-shot2019.03 | — | — | — | — | — | — | — | — | — | — | 61.3 | — | — | — | — | |
| OSLSMShots=5-shot2019.03 | — | — | — | — | — | — | — | — | — | — | 61.5 | — | — | — | — | |
| OSLSMShot=1-shot2019.08 | — | — | — | — | — | — | — | — | — | — | — | 61.3 | — | — | — | |
| OSLSMShot=5-shot2019.08 | — | — | — | — | — | — | — | — | — | — | — | 61.5 | — | — | — | |
| PANetShot=1-shot2019.08 | — | — | — | — | — | — | — | — | — | — | — | 66.5 | — | — | — | |
| PANetShot=5-shot2019.08 | — | — | — | — | — | — | — | — | — | — | — | 70.7 | — | — | — | |
| PANet-initShot=1-shot2019.08 | — | — | — | — | — | — | — | — | — | — | — | 58.9 | — | — | — | |
| PANet-initShot=5-shot2019.08 | — | — | — | — | — | — | — | — | — | — | — | 65.7 | — | — | — | |
| PFENetFPS=15.7, Params=34.3 M2021.03 | — | — | — | — | 60.8 | — | — | — | — | — | — | — | — | — | — | |
| PFENetvis. backb.=RN502021.12 | — | — | — | — | 60.2 | — | — | — | — | — | — | 73.3 | — | — | — | |
| PFENetvis. backb.=RN1012021.12 | — | — | — | — | 59.6 | — | — | — | — | — | — | 72.9 | — | — | — | |
| PLShots=1-shot2019.03 | — | — | — | — | — | — | — | — | — | — | 61.2 | — | — | — | — | |
| PLShots=5-shot2019.03 | — | — | — | — | — | — | — | — | — | — | 62.3 | — | — | — | — | |
| PLShot=1-shot2019.08 | — | — | — | — | — | — | — | — | — | — | — | 61.2 | — | — | — | |
| PLShot=5-shot2019.08 | — | — | — | — | — | — | — | — | — | — | — | 62.3 | — | — | — | |
| PMMsFPS=18.2, Params=19.6 M2021.03 | — | — | — | — | 56.3 | — | — | — | — | — | — | — | — | — | — | |
| PPNetvis. backb.=RN502021.12 | — | — | — | — | 52.8 | — | — | — | — | — | — | 69.2 | — | — | — | |
| PPNetvis. backb.=RN1012021.12 | — | — | — | — | 55.2 | — | — | — | — | — | — | 70.9 | — | — | — | |
| RePRIvis. backb.=RN502021.12 | — | — | — | — | 59.7 | — | — | — | — | — | — | — | — | — | — | |
| RePRIvis. backb.=RN1012021.12 | — | — | — | — | 59.4 | — | — | — | — | — | — | — | — | — | — | |
| SG-OneShot=1-shot2019.08 | — | — | — | — | — | — | — | — | — | — | — | 63.9 | — | — | — | |
| SG-OneShot=5-shot2019.08 | — | — | — | — | — | — | — | — | — | — | — | 65.9 | — | — | — | |
| ViTSeg (PC+)t=0.2, vis. backb.=ViT (IN), Trained on=PhraseCut+2021.12 | — | — | — | — | 39 | — | — | — | — | — | — | 59 | — | 62.4 | — |