Metric: Validation mIoU (higher is better)
| # | Model↕ | Validation mIoU▼ | Extra Data | Paper | Date↕ | Code |
|---|---|---|---|---|---|---|
| 1 | UniMatch V2 (DINOv2-B) | 83.6 | No | UniMatch V2: Pushing the Limit of Semi-Supervise... | 2024-10-14 | Code |
| 2 | SemiVL (ViT-B/16) | 77.9 | No | SemiVL: Semi-Supervised Semantic Segmentation wi... | 2023-11-27 | Code |
| 3 | CorrMatch (Deeplabv3+ with ResNet-101) | 77.3 | No | CorrMatch: Label Propagation via Correlation Mat... | 2023-06-07 | Code |
| 4 | FARCLUSS | 77.2 | No | FARCLUSS: Fuzzy Adaptive Rebalancing and Contras... | 2025-06-11 | Code |
| 5 | S4MC | 77 | No | Semi-Supervised Semantic Segmentation via Margin... | 2023-08-26 | Code |
| 6 | Dual Teacher (DeepLab v3+ with ResNet-101 pretraind on ImageNet-1K) | 76.81 | No | - | - | Code |
| 7 | UniMatch (DeepLab v3+ with ResNet-101 pretraind on ImageNet-1K) | 76.59 | No | Revisiting Weak-to-Strong Consistency in Semi-Su... | 2022-08-21 | Code |
| 8 | n-CPS (ResNet-50) | 76.08 | No | n-CPS: Generalising Cross Pseudo Supervision to ... | 2021-12-14 | - |
| 9 | CW-BASS (DeepLab v3+ with ResNet-50) | 75 | No | Confidence-Weighted Boundary-Aware Learning for ... | 2025-02-21 | Code |
| 10 | CPS (DeepLab v3+ with ResNet-101) | 69.8 | No | Semi-Supervised Semantic Segmentation with Cross... | 2021-06-02 | Code |