Cross-modal Geo-localization on CVG-Text Tokyo
43.42Recall@1PLANET
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| PLANET2026.03 | 43.42 | 46.42 | |
| CrossText2Loc2026.03 | 41.75 | 43.86 | |
| Long-CLIP2026.03 | 33.83 | 35.5 | |
| EVA2-CLIP-L/142026.03 | 31 | 34.08 | |
| BLIP2026.03 | 29.75 | 33.67 | |
| SigLIP-SO400M2026.03 | 28.42 | 31.5 | |
| CLIP-L/142026.03 | 28.08 | 30.5 | |
| RemoteCLIP2026.03 | 26.42 | 29.08 | |
| TAG-CLIP2026.03 | 24.33 | 28.08 | |
| CLIP-B/162026.03 | 24 | 27.25 | |
| EVA2-CLIP-B/162026.03 | 22.5 | 25.25 | |
| X2-VLM2026.03 | 17.83 | 20.75 | |
| SigLIP-B/162026.03 | 15.58 | 17.92 | |
| X-VLM2026.03 | 12.46 | 14.34 | |
| ViLT2026.03 | 10.83 | 15.5 |