Semantic Segmentation on ADE20K
Metric: Mean IoU (class) (higher is better)
Metric: Mean IoU (class) (higher is better)
| # | Model↕ | Mean IoU (class)▼ | Extra Data | Paper | Date↕ | Code |
|---|---|---|---|---|---|---|
| 1 | EoMT (DINOv2-L, single-scale, 512x512) | 58.4 | No | Your ViT is Secretly an Image Segmentation Model | 2025-03-24 | Code |
| 2 | FastViT-MA36 | 44.6 | No | FastViT: A Fast Hybrid Vision Transformer using ... | 2023-03-24 | Code |
| 3 | FastViT-SA36 | 42.9 | No | FastViT: A Fast Hybrid Vision Transformer using ... | 2023-03-24 | Code |
| 4 | FastViT-SA24 | 41 | No | FastViT: A Fast Hybrid Vision Transformer using ... | 2023-03-24 | Code |
| 5 | FastViT-SA12 | 38 | No | FastViT: A Fast Hybrid Vision Transformer using ... | 2023-03-24 | Code |