Image-to-Text Retrieval on LIT-CN
87.6Recall@1FG-CLIP 2
Evaluation results
| Method | Links | |
|---|---|---|
| FG-CLIP 2Backbone=ViT-So/162025.10 | 87.6 | |
| FG-CLIP 2Backbone=ViT-L/162025.10 | 86.3 | |
| FG-CLIP 2Backbone=ViT-B/162025.10 | 82.4 | |
| Meta CLIP 2Backbone=ViT-H/142025.10 | 77.2 | |
| Chinese-CLIPBackbone=ViT-L/142025.10 | 48.6 | |
| R2D2Backbone=ViT-L/142025.10 | 48.3 | |
| Chinese-CLIPBackbone=ViT-B/162025.10 | 45.7 | |
| R2D2Backbone=ViT-B/162025.10 | 35.7 | |
| SigLIP 2Backbone=ViT-So/162025.10 | 17 | |
| SigLIP 2Backbone=ViT-L/162025.10 | 14.8 | |
| SigLIP 2Backbone=ViT-B/162025.10 | 4.6 |