Remote Sensing Image-Text Retrieval on RRQC Distance 4 1.0 (test)
9.96RRQCR@k,4RUNE
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| RUNEImage Encoder=YOLO11, Text-encoder=GPT-4o2025.12 | 9.96 | -5.39 | |
| RemoteCLIPImage Encoder=RN50, Text-encoder=Transformer2025.12 | 7.8 | -6.82 | |
| RemoteCLIPImage Encoder=ViT-B-32, Text-encoder=Transformer2025.12 | 7.06 | -8.34 | |
| RemoteCLIPImage Encoder=ViT-L-14, Text-encoder=Transformer2025.12 | 5.86 | -5.45 | |
| RS-LLaVAImage Encoder=CLIP-ViT-L-336px, Text-encoder=Mistral-7B2025.12 | -4.88 | -15.82 | |
| CLIPImage Encoder=ViT-L-14, Text-encoder=Transformer2025.12 | -10.02 | -17.15 | |
| CLIPImage Encoder=RN50, Text-encoder=Transformer2025.12 | -11.08 | -21.41 | |
| GeoRSCLIPImage Encoder=ViT-B-32, Text-encoder=Transformer2025.12 | -14.74 | -21.55 | |
| OpenCLIPImage Encoder=ViT-B-32, Text-encoder=Transformer2025.12 | -15.39 | -22 | |
| CLIPImage Encoder=ViT-B-32, Text-encoder=Transformer2025.12 | -15.39 | -22 | |
| LLaVA-1.6Image Encoder=CLIP-ViT-L-336px, Text-encoder=Mistral-7b-HF2025.12 | -20.13 | -29.65 |