Metric: mIoU (test) (higher is better)
| # | Model↕ | mIoU (test)▼ | Extra Data | Paper | Date↕ | Code |
|---|---|---|---|---|---|---|
| 1 | SERE (ViT-B/16, 100ep, 224x224, SSL+FT) | 63.3 | No | SERE: Exploring Feature Self-relation for Self-s... | 2022-06-10 | Code |
| 2 | TEC (ViT-B/16, 224x224, SSL+FT, mmseg) | 62.5 | No | Towards Sustainable Self-supervised Learning | 2022-10-20 | Code |
| 3 | MAE (ViT-B/16, 224x224, SSL+FT, mmseg) | 61.2 | No | Masked Autoencoders Are Scalable Vision Learners | 2021-11-11 | Code |
| 4 | MAE (ViT-B/16, 224x224, SSL+FT) | 60.2 | No | Masked Autoencoders Are Scalable Vision Learners | 2021-11-11 | Code |
| 5 | SERE (ViT-S/16, 100ep, 224x224, SSL+FT, mmseg) | 59 | No | SERE: Exploring Feature Self-relation for Self-s... | 2022-06-10 | Code |
| 6 | SERE (ViT-S/16, 100ep, 224x224, SSL+FT) | 57.8 | No | SERE: Exploring Feature Self-relation for Self-s... | 2022-06-10 | Code |
| 7 | RF-ConvNext-Tiny (rfmerge, P4, 224x224, SUP) | 51.1 | No | RF-Next: Efficient Receptive Field Search for Co... | 2022-06-14 | Code |
| 8 | RF-ConvNext-Tiny (rfmultiple, P4, 224x224, SUP) | 50.5 | No | RF-Next: Efficient Receptive Field Search for Co... | 2022-06-14 | Code |
| 9 | RF-ConvNext-Tiny (rfsingle, P4, 224x224, SUP) | 50.5 | No | RF-Next: Efficient Receptive Field Search for Co... | 2022-06-14 | Code |
| 10 | ConvNext-Tiny (P4, 224x224, SUP) | 48.8 | No | A ConvNet for the 2020s | 2022-01-10 | Code |
| 11 | SERE (ViT-B/16, 100ep, 224x224, SSL) | 48.2 | No | SERE: Exploring Feature Self-relation for Self-s... | 2022-06-10 | Code |
| 12 | TEC (ViT-B/16, 224x224, SSL, mmseg) | 46 | No | Towards Sustainable Self-supervised Learning | 2022-10-20 | Code |
| 13 | SERE (ViT-S/16, 100ep, 224x224, SSL, mmseg) | 40.5 | No | SERE: Exploring Feature Self-relation for Self-s... | 2022-06-10 | Code |
| 14 | MAE (ViT-B/16, 224x224, SSL, mmseg) | 40.3 | No | Masked Autoencoders Are Scalable Vision Learners | 2021-11-11 | Code |
| 15 | SERE (ViT-S/16, 100ep, 224x224, SSL) | 40.2 | No | SERE: Exploring Feature Self-relation for Self-s... | 2022-06-10 | Code |
| 16 | MAE (ViT-B/16, 224x224, SSL) | 37 | No | Masked Autoencoders Are Scalable Vision Learners | 2021-11-11 | Code |
| 17 | PASS (ResNet-50 D16, 224x224, LUSS) | 20.8 | No | Large-scale Unsupervised Semantic Segmentation | 2021-06-06 | Code |
| 18 | PASS (ResNet-50 D32, 224x224, LUSS) | 20.3 | No | Large-scale Unsupervised Semantic Segmentation | 2021-06-06 | Code |
| 19 | PASS | 11 | No | Large-scale Unsupervised Semantic Segmentation | 2021-06-06 | Code |