Image Classification on RSI-CB256
61.8AccuracyT-GeoRSCLIP
Evaluation Results
| Method | Links | |
|---|---|---|
| T-GeoRSCLIPSetting=Transductive, Transductive Inference (TransCLIP)=true2025.06 | 61.8 | |
| SOTASetting=Transductive, Transductive Inference (TransCLIP)=false2025.06 | 61.1 | |
| T-SkyCLIP50Setting=Transductive, Transductive Inference (TransCLIP)=true2025.06 | 59.1 | |
| GeoRSCLIPArchitecture=ViT-L/14, Setting=Zero-shot, Template=DST2026.06 | 55.7 | |
| GeoRSCLIPARCH=ViT-L/14, Zero-shot=true, Approach=Meta-prompting (LLaMA-3.1)2026.06 | 53.5 | |
| SkyCLIPArchitecture=ViT-L/14, Setting=Zero-shot, Template=DST2026.06 | 53.1 | |
| RemoteCLIPBackbone=ViT-L-14, zero-shot=true2023.06 | 52.82 | |
| RemoteCLIPBackbone=ViT-L-14, zero-shot=true2023.06 | 52.82 | |
| RemoteCLIPBackbone=ViT-L, zero-shot=true2025.12 | 51.55 | |
| RemoteCLIPARCH=ViT-L/14, Zero-shot=true, Approach=Meta-prompting (LLaMA-3.1)2026.06 | 51 | |
| SkyCLIPARCH=ViT-L/14, Zero-shot=true, Approach=Meta-prompting (LLaMA-3.1)2026.06 | 50.8 | |
| CLIP LAION-RSArchitecture=ViT-L/14, Setting=Zero-shot, Template=DST2026.06 | 50.4 | |
| RemoteCLIPArchitecture=ViT-L/14, Setting=Zero-shot, Template=DST2026.06 | 49.3 | |
| T-RemoteCLIPSetting=Transductive, Transductive Inference (TransCLIP)=true2025.06 | 49.2 | |
| CLIP LAION-RSARCH=ViT-L/14, Zero-shot=true, Approach=Meta-prompting (LLaMA-3.1)2026.06 | 48.9 | |
| CLIPArchitecture=ViT-L/14, Setting=Zero-shot, Template=DST2026.06 | 48.9 | |
| SigLIPARCH=ViT-L/16, Zero-shot=true, Approach=Meta-prompting (LLaMA-3.1)2026.06 | 48.7 | |
| GeoRSCLIPARCH=ViT-B/32, Zero-shot=true, Approach=Meta-prompting (LLaMA-3.1)2026.06 | 48.6 | |
| CLIPBackbone=ViT-L-14, zero-shot=true2023.06 | 47.94 | |
| CLIPBackbone=ViT-L-14, zero-shot=true2023.06 | 47.94 | |
| SigLIP2ARCH=ViT-L/16, Zero-shot=true, Approach=Meta-prompting (LLaMA-3.1)2026.06 | 47.9 | |
| SkyCLIP50Setting=Transductive, Transductive Inference (TransCLIP)=false2025.06 | 47.1 | |
| GeoRSCLIPSetting=Transductive, Transductive Inference (TransCLIP)=false2025.06 | 46.5 | |
| GeoMELTBackbone=BEiT-3, zero-shot=true2025.12 | 46.45 | |
| T-CLIPSetting=Transductive, Transductive Inference (TransCLIP)=true2025.06 | 46.4 | |
| GeoRSCLIPArchitecture=ViT-B/32, Setting=Zero-shot, Template=DST2026.06 | 46.4 | |
| SigLIP2ARCH=ViT-B/16, Zero-shot=true, Approach=Meta-prompting (LLaMA-3.1)2026.06 | 44.8 | |
| MetaCLIP-400MArchitecture=ViT-L/14, Setting=Zero-shot, Template=DST2026.06 | 44.8 | |
| SigLIP2Architecture=ViT-L/16, Setting=Zero-shot, Template=DST2026.06 | 44.4 | |
| MetaCLIP-2.5BArchitecture=ViT-L/14, Setting=Zero-shot, Template=DST2026.06 | 44 | |
| CLIPBackbone=ViT-L, zero-shot=true2025.12 | 43.65 | |
| CLIPArchitecture=ViT-B/16, Setting=Zero-shot, Template=DST2026.06 | 43 | |
| SigLIP2Architecture=ViT-B/16, Setting=Zero-shot, Template=DST2026.06 | 42.7 | |
| RemoteCLIPBackbone=ViT-B, zero-shot=true2025.12 | 42.56 | |
| MetaCLIP-400MARCH=ViT-B/16, Zero-shot=true, Approach=Meta-prompting (LLaMA-3.1)2026.06 | 41.7 | |
| RemoteCLIPSetting=Transductive, Transductive Inference (TransCLIP)=false2025.06 | 41.5 | |
| SigLIPARCH=ViT-B/16, Zero-shot=true, Approach=Meta-prompting (LLaMA-3.1)2026.06 | 41.5 | |
| SigLIPArchitecture=ViT-L/16, Setting=Zero-shot, Template=DST2026.06 | 41.4 | |
| CLIPARCH=ViT-L/14, Zero-shot=true, Approach=Meta-prompting (LLaMA-3.1)2026.06 | 41.2 | |
| MetaCLIP-2.5BArchitecture=ViT-B/16, Setting=Zero-shot, Template=DST2026.06 | 41.1 | |
| MetaCLIP-400MARCH=ViT-L/14, Zero-shot=true, Approach=Meta-prompting (LLaMA-3.1)2026.06 | 41 | |
| CLIPSetting=Transductive, Transductive Inference (TransCLIP)=false2025.06 | 40.3 | |
| MetaCLIP-2.5BARCH=ViT-B/16, Zero-shot=true, Approach=Meta-prompting (LLaMA-3.1)2026.06 | 40 | |
| RemoteCLIPArchitecture=ViT-B/32, Setting=Zero-shot, Template=DST2026.06 | 39.7 | |
| RemoteCLIPBackbone=ViT-B-32, zero-shot=true2023.06 | 39.5 | |
| RemoteCLIPBackbone=ViT-B-32, zero-shot=true2023.06 | 39.5 | |
| MetaCLIP-400MArchitecture=ViT-B/16, Setting=Zero-shot, Template=DST2026.06 | 39.4 | |
| MetaCLIP-2.5BARCH=ViT-L/14, Zero-shot=true, Approach=Meta-prompting (LLaMA-3.1)2026.06 | 39 | |
| RemoteCLIPARCH=ViT-B/32, Zero-shot=true, Approach=Meta-prompting (LLaMA-3.1)2026.06 | 39 | |
| CLIPBackbone=ViT-B-32, zero-shot=true2023.06 | 37.35 | |
| CLIPBackbone=ViT-B-32, zero-shot=true2023.06 | 37.35 | |
| CLIPARCH=ViT-B/16, Zero-shot=true, Approach=Meta-prompting (LLaMA-3.1)2026.06 | 37.2 | |
| SigLIPArchitecture=ViT-B/16, Setting=Zero-shot, Template=DST2026.06 | 36.8 | |
| RemoteCLIPArchitecture=RN-50, Setting=Zero-shot, Template=DST2026.06 | 35 | |
| CLIPBackbone=ResNet-50, zero-shot=true2023.06 | 34.24 | |
| CLIPBackbone=ResNet-50, zero-shot=true2023.06 | 34.24 | |
| RemoteCLIPBackbone=ResNet-50, zero-shot=true2023.06 | 33.03 | |
| RemoteCLIPBackbone=ResNet-50, zero-shot=true2023.06 | 33.03 | |
| RemoteCLIPARCH=RN-50, Zero-shot=true, Approach=Meta-prompting (LLaMA-3.1)2026.06 | 31.8 | |
| CLIPBackbone=ViT-B, zero-shot=true2025.12 | 29.4 |