| Semi-Supervised Semantic Segmentation via Marginal Contextual Information | ✓ Link | 81.11 | S4MC | 2023-08-26 |
| Semi-supervised Semantic Segmentation with Prototype-based Consistency Regularization | ✓ Link | 80.91% | PCR (DeepLab v3+ with ResNet-101 pretraind on ImageNet-1K) | 2022-10-10 |
| Semi-Supervised Semantic Segmentation Using Unreliable Pseudo-Labels | ✓ Link | 80.5% | U2PL (DeepLab v3+ with ResNet-101 pretraind on ImageNet-1K, CutMix) | 2022-03-08 |
| Semi-Supervised Semantic Segmentation via Adaptive Equalization Learning | ✓ Link | 80.29% | AEL (DeepLab v3+ with ResNet-101 pretraind on ImageNet-1K) | 2021-10-11 |
| n-CPS: Generalising Cross Pseudo Supervision to n Networks for Semi-Supervised Semantic Segmentation | | 80.26% | n-CPS (ResNet-101) | 2021-12-14 |
| Perturbed and Strict Mean Teachers for Semi-supervised Semantic Segmentation | ✓ Link | 79.76% | PS-MT
(DeepLab v3+ with ImageNet-pretrained ResNet-101, single scale inference) | 2021-11-25 |
| GuidedMix-Net: Learning to Improve Pseudo Masks Using Labeled Images as Reference | ✓ Link | 78.2% | GuidedMix-Net(DeepLab v2 with ResNet101, input-size: 512x512 with multi-scale and flip, ImageNet pretrained) | 2021-06-29 |
| Conservative-Progressive Collaborative Learning for Semi-supervised Semantic Segmentation | ✓ Link | 77.67% | CPCL (DeepLab v3+ with ResNet-101) | 2022-11-30 |
| Learning Pseudo Labels for Semi-and-Weakly Supervised Semantic Segmentation | ✓ Link | 77.26% | PCT (DeepLab v3+ with ResNet-50 pretrained on ImageNet-1K) | 2022-08-02 |
| n-CPS: Generalising Cross Pseudo Supervision to n Networks for Semi-Supervised Semantic Segmentation | | 77.07% | n-CPS (ResNet-50) | 2021-12-14 |
| GuidedMix-Net: Learning to Improve Pseudo Masks Using Labeled Images as Reference | ✓ Link | 76.5% | GuidedMix-Net(DeepLab v2 with ResNet101, ImageNet pretrained) | 2021-06-29 |
| Conservative-Progressive Collaborative Learning for Semi-supervised Semantic Segmentation | ✓ Link | 75.3% | CPCL (DeepLab v3+ with ResNet-50) | 2022-11-30 |
| Dense FixMatch: a simple semi-supervised learning method for pixel-wise prediction tasks | ✓ Link | 74.73% | Dense FixMatch (DeepLabv3+ ResNet-101, over-sampling, single pass eval) | 2022-10-18 |
| Dense FixMatch: a simple semi-supervised learning method for pixel-wise prediction tasks | ✓ Link | 71.69% | Dense FixMatch (DeepLabv3+ ResNet-50, over-sampling, single pass eval) | 2022-10-18 |