Medical Image Segmentation on CVC-ClinicDB (test)
95.43DiceS2M-Net
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| S2M-NetLoss=MASL2026.01 | 95.43 | — | — | — | — | |
| PVT-EMCAD-B2#Params=26.76M, #FLOPs=5.6G, Resolution=256x2562024.05 | 95.21 | — | — | — | — | |
| PVT-EMCAD-B0#Params=3.92M, #FLOPs=0.84G, Resolution=256x2562024.05 | 94.6 | — | — | — | — | |
| PVT-CASCADE#Params=34.12M, #FLOPs=7.62G, Resolution=256x2562024.05 | 94.53 | — | — | — | — | |
| SSFormerVariant=L2022.03 | 94.47 | 89.95 | — | — | — | |
| HadBalanceBackbone=nnUNet2026.06 | 94.41 | 89.85 | — | — | 75.07 | |
| MSRF-Net2022.03 | 94.2 | 90.43 | — | — | — | |
| SSFormer-L#Params=66.22M, #FLOPs=17.28G, Resolution=256x2562024.05 | 94.18 | — | — | — | — | |
| UACANet-L#Params=69.16M, #FLOPs=31.51G, Resolution=256x2562024.05 | 94.16 | — | — | — | — | |
| PolypPVT#Params=25.11M, #FLOPs=5.30G, Resolution=256x2562024.05 | 94.13 | — | — | — | — | |
| CaraNet#Params=46.64M, #FLOPs=11.48G, Resolution=256x2562024.05 | 94.08 | — | — | — | — | |
| TransUNet#Params=105.32M, #FLOPs=38.52G, Resolution=256x2562024.05 | 93.9 | — | — | — | — | |
| LatentFMApproach=Generative Segmentation Methods2025.12 | 93.71 | 88.16 | — | — | — | |
| TransFuse#Params=143.74M, #FLOPs=82.71G, Resolution=256x2562024.05 | 93.62 | — | — | — | — | |
| SwinUNet2026.06 | 93.54 | 88.21 | — | — | 66.2 | |
| NASHBackbone=nnUNet2026.06 | 93.52 | 88.14 | — | — | 65.86 | |
| FairGradBackbone=nnUNet2026.06 | 93.49 | 88.06 | — | — | 65.05 | |
| nnUNet2026.06 | 93.38 | 88.55 | — | — | 73.46 | |
| LSBackbone=nnUNet2026.06 | 93.37 | 88.35 | — | — | 73.43 | |
| DeepLabv3+#Params=39.76M, #FLOPs=14.92G, Resolution=256x2562024.05 | 93.24 | — | — | — | — | |
| S2M-NetLoss=standard Dice loss2026.01 | 92.95 | — | — | — | — | |
| SSFormerVariant=S2022.03 | 92.68 | 87.59 | — | — | — | |
| ACC-UNetparams=16.8M, FLOPS=38G2023.08 | 92.67 | — | — | — | — | |
| UCTransnetparams=66.4M, FLOPS=38.8G2023.08 | 92.57 | — | — | — | — | |
| SwinUNet#Params=27.17M, #FLOPs=6.2G, Resolution=224x2242024.05 | 92.42 | — | — | — | — | |
| Attention-UNet2026.06 | 92.31 | 86.77 | — | — | 69.86 | |
| AttnUNet#Params=34.88M, #FLOPs=66.64G, Resolution=256x2562024.05 | 92.2 | — | — | — | — | |
| UNet++#Params=9.16M, #FLOPs=34.65G, Resolution=256x2562024.05 | 92.17 | — | — | — | — | |
| UNet2026.06 | 92.12 | 86.51 | — | — | 68.56 | |
| UNet#Params=24.53M, #FLOPs=65.53G, Resolution=256x2562024.05 | 92.11 | — | — | — | — | |
| UNet++2026.06 | 92.04 | 86.37 | — | — | 68.54 | |
| UMambaLoss=MASL2026.01 | 91.89 | — | — | — | — | |
| TransUNetApproach=Deterministic Segmentation Methods2025.12 | 91.8 | 85.9 | — | — | — | |
| TransUNet2026.06 | 91.74 | 86.15 | — | — | 66.96 | |
| PraNet#Params=32.55M, #FLOPs=6.93G, Resolution=256x2562024.05 | 91.71 | — | — | — | — | |
| U-Net2022.03 | 91.45 | 86.54 | — | — | — | |
| DuckNetLoss=MASL2026.01 | 91.23 | — | — | — | — | |
| TopoMamba-2DParam(M)=23.92026.04 | 91.05 | 82.37 | — | 98.91 | — | |
| SwinUNetLoss=MASL2026.01 | 90.78 | — | — | — | — | |
| Swin-Unetparams=27.2M, FLOPS=6.2G2023.08 | 90.69 | — | — | — | — | |
| UNetparams=14M, FLOPS=37G2023.08 | 90.66 | — | — | — | — | |
| Swin-UMambaParam(M)=27.62026.04 | 90.5 | 80.4 | — | 97.33 | — | |
| PraNetLoss=MASL2026.01 | 90.45 | — | — | — | — | |
| TransUNetLoss=MASL2026.01 | 90.34 | — | — | — | — | |
| UMambaLoss=standard Dice loss2026.01 | 90.23 | — | — | — | — | |
| UNeXt#Params=1.47M, #FLOPs=0.57G, Resolution=256x2562024.05 | 90.2 | — | — | — | — | |
| VM-UNetParam(M)=31.02026.04 | 89.84 | 79.62 | — | 96.36 | — | |
| DuckNetLoss=standard Dice loss2026.01 | 89.78 | — | — | — | — | |
| RAPUNetLoss=MASL2026.01 | 89.67 | — | — | — | — | |
| SMESwin-Unetparams=169.8M, FLOPS=6.4G2023.08 | 89.62 | — | — | — | — | |
| LatentDMApproach=Generative Segmentation Methods2025.12 | 89.56 | 81.11 | — | — | — | |
| Mamba-UNetParam(M)=19.12026.04 | 89.4 | 79.46 | — | 96.24 | — | |
| U-Net++Loss=MASL2026.01 | 89.12 | — | — | — | — | |
| SwinUNetLoss=standard Dice loss2026.01 | 89.12 | — | — | — | — | |
| FMApproach=Generative Segmentation Methods2025.12 | 89.02 | 81 | — | — | — | |
| Deeplabv3+2022.03 | 88.97 | 87.06 | — | — | — | |
| U-MambaParam(M)=23.422026.04 | 88.74 | 79.08 | — | 96.3 | — | |
| PraNetLoss=standard Dice loss2026.01 | 88.67 | — | — | — | — | |
| TransUNetLoss=standard Dice loss2026.01 | 88.67 | — | — | — | — | |
| MultiResUNetparams=7.3M, FLOPS=1.1G2023.08 | 88.2 | — | — | — | — | |
| U-NetLoss=MASL2026.01 | 87.54 | — | — | — | — | |
| RAPUNetLoss=standard Dice loss2026.01 | 87.45 | — | — | — | — | |
| U-Net++Loss=standard Dice loss2026.01 | 86.89 | — | — | — | — | |
| SwinUNetParam(M)=27.22026.04 | 86.67 | 79.21 | — | 95.67 | — | |
| UNetParam(M)=31.02026.04 | 86.12 | 78.02 | — | 95.46 | — | |
| U-NetLoss=standard Dice loss2026.01 | 85.23 | — | — | — | — | |
| U-Net++2022.03 | 84.53 | 75.59 | — | — | — | |
| UNetApproach=Deterministic Segmentation Methods2025.12 | 83.13 | 74.69 | — | — | — | |
| DMApproach=Generative Segmentation Methods2025.12 | 82.44 | 73.13 | — | — | — | |
| nnUNetApproach=Deterministic Segmentation Methods2025.12 | 81.3 | 73.3 | — | — | — | |
| UNet++Approach=Deterministic Segmentation Methods2025.12 | 80.16 | 72.36 | — | — | — | |
| Certified Medical Image Segmentation with Diffusion ModelsModel=ResUNet++, σ=0.25, R=0.172023.10 | 65 | 57 | 4 | — | — | |
| SEGCERTIFYModel=ResUNet++, σ=0.25, R=0.172023.10 | 63 | 56 | 5 | — | — | |
| Certified Medical Image Segmentation with Diffusion ModelsModel=ResUNet++, σ=0.50, R=0.342023.10 | 45 | 39 | 7 | — | — | |
| Certified Medical Image Segmentation with Diffusion ModelsModel=ResUNet++, σ=1.00, R=0.672023.10 | 26 | 23 | 14 | — | — | |
| SEGCERTIFYModel=ResUNet++, σ=0.50, R=0.342023.10 | 15 | 10 | 1 | — | — | |
| SEGCERTIFYModel=ResUNet++, σ=1.00, R=0.672023.10 | 0 | 0 | 0 | — | — |