Semantic Segmentation on NYC 0.3 (train)
61.88mIoUscaleMAE (3B)
Evaluation results
| Method | Links | |
|---|---|---|
| scaleMAE (3B)Backbone=ViT-Large, Segmentation Framework=UperNet, Parameter Count=3B, Evaluation Protocol=Fine-tuned2025.11 | 61.88 | |
| satMAE (3B)Backbone=ViT-Large, Segmentation Framework=UperNet, Parameter Count=3B, Evaluation Protocol=Fine-tuned2025.11 | 61.01 | |
| DOFABackbone=ViT-Large, Segmentation Framework=UperNet, Evaluation Protocol=Fine-tuned2025.11 | 60.22 | |
| satMAE++ (3B)Backbone=ViT-Large, Segmentation Framework=UperNet, Parameter Count=3B, Evaluation Protocol=Fine-tuned2025.11 | 59.57 | |
| LandSegmenterBackbone=SAM2+DOFA, Evaluation Protocol=Fine-tuned2025.11 | 59.41 | |
| SAM2+HR+DOFA*Backbone=SAM2+DOFA, Evaluation Protocol=Fine-tuned2025.11 | 57.45 | |
| SAM2+HR (3B)Backbone=SAM2, Parameter Count=3B, Evaluation Protocol=Fine-tuned2025.11 | 53.64 | |
| SAM2 (3B)Backbone=SAM2, Parameter Count=3B, Evaluation Protocol=Fine-tuned2025.11 | 43.49 |