Image Classification on fMoW
81.7AccuracyRS-Global MTP
Evaluation results
| Method | Links | ||
|---|---|---|---|
| RS-Global MTPTraining Mode=Fine-tuning2025.10 | 81.7 | — | |
| SOTA RS foundation modelTraining Mode=Fine-tuning2025.10 | 79.3 | — | |
| STAM (DeiT-Small (default))Attention used complete image=No, Classification used complete image=No, Evaluation Criterion=Equivalent sensing to PatchDrop, Time steps (t)=182022.04 | 71.5 | — | |
| STAM (DeiT-Small (default))Attention used complete image=No, Classification used complete image=No, Evaluation Criterion=Equivalent accuracy to PatchDrop, Time steps (t)=102022.04 | 68.8 | — | |
| PatchDropAttention used complete image=Yes, Classification used complete image=Yes2022.04 | 68.3 | — | |
| DINOv2Architecture=ViT-g/14, Pre-training Data=LVD, Resolution=518x518, Evaluation Protocol=Attentive Probing, Backbone State=Frozen2024.01 | 66.7 | — | |
| AIMv2architecture=ViT-3B/14, Resolution=448px2024.11 | 66.1 | — | |
| STNAttention used complete image=Yes, Classification used complete image=Yes2022.04 | 64.8 | — | |
| AIM-7B†Architecture=ViT-7B/14, Pre-training Data=DFN-2B+, Feature Extraction Layer=20th, Evaluation Protocol=Attentive Probing, Backbone State=Frozen2024.01 | 63.5 | — | |
| AIM-7BArchitecture=ViT-7B/14, Pre-training Data=DFN-2B+, Evaluation Protocol=Attentive Probing, Backbone State=Frozen2024.01 | 61.7 | — | |
| AIM-3BArchitecture=ViT-3B/14, Pre-training Data=DFN-2B+, Evaluation Protocol=Attentive Probing, Backbone State=Frozen2024.01 | 60.9 | — | |
| iBOTArchitecture=ViT-L/16, Pre-training Data=IN-21k, Evaluation Protocol=Attentive Probing, Backbone State=Frozen2024.01 | 60.3 | — | |
| MAEArchitecture=ViT-2B/14, Pre-training Data=IG-3B, Evaluation Protocol=Attentive Probing, Backbone State=Frozen2024.01 | 60.1 | — | |
| AIM-1BArchitecture=ViT-1B/14, Pre-training Data=DFN-2B+, Evaluation Protocol=Attentive Probing, Backbone State=Frozen2024.01 | 59.8 | — | |
| CROMABackbone=ViT-L2026.03 | 59 | — | |
| GeoSANEBackbone=ViT-L2026.03 | 58.9 | — | |
| AIM-0.6BArchitecture=ViT-H/14, Pre-training Data=DFN-2B+, Evaluation Protocol=Attentive Probing, Backbone State=Frozen2024.01 | 58.3 | — | |
| SatMAEBackbone=ViT-L2026.03 | 58.2 | — | |
| DINOArchitecture=ViT-B/8, Pre-training Data=IN-1k, Evaluation Protocol=Attentive Probing, Backbone State=Frozen2024.01 | 57 | — | |
| MAEArchitecture=ViT-H/14, Pre-training Data=IN-1k, Evaluation Protocol=Attentive Probing, Backbone State=Frozen2024.01 | 56.9 | — | |
| BEiTArchitecture=ViT-L/14, Pre-training Data=IN-21k, Evaluation Protocol=Attentive Probing, Backbone State=Frozen2024.01 | 47.7 | — |