Image-to-Text Retrieval on DCI-CN
62.7Recall@1FG-CLIP 2
Evaluation results
| Method | Links | |
|---|---|---|
| FG-CLIP 2Backbone=ViT-So/162025.10 | 62.7 | |
| FG-CLIP 2Backbone=ViT-L/162025.10 | 60.4 | |
| FG-CLIP 2Backbone=ViT-B/162025.10 | 53.9 | |
| Meta CLIP 2Backbone=ViT-H/142025.10 | 53.8 | |
| R2D2Backbone=ViT-L/142025.10 | 35.6 | |
| Chinese-CLIPBackbone=ViT-L/142025.10 | 31.4 | |
| Chinese-CLIPBackbone=ViT-B/162025.10 | 30.1 | |
| R2D2Backbone=ViT-B/162025.10 | 25.9 | |
| SigLIP 2Backbone=ViT-L/162025.10 | 13.6 | |
| SigLIP 2Backbone=ViT-So/162025.10 | 13.4 | |
| SigLIP 2Backbone=ViT-B/162025.10 | 5 |