Text-to-Image Retrieval on Wukong (test)
68.9R@1CLIP_ViT-L
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| CLIP_ViT-LBackbone=ViT-L, Evaluation Protocol=Fine-tuned2022.02 | 68.9 | 98 | 99.8 | 89.6 | |
| Wukong_ViT-LBackbone=ViT-L, Evaluation Protocol=Fine-tuned2022.02 | 63.8 | 93 | 97.3 | 85.8 | |
| CLIP_ViT-BBackbone=ViT-B, Evaluation Protocol=Fine-tuned2022.02 | 53.1 | 85.4 | 92.6 | 78.6 | |
| CLIP_Swin-LBackbone=Swin-L, Evaluation Protocol=Fine-tuned2022.02 | 51 | 83.4 | 90.9 | 76.7 | |
| FILIP_ViT-LBackbone=ViT-L, Evaluation Protocol=Fine-tuned2022.02 | 50.4 | 84.1 | 92 | 76.3 | |
| Wukong_Swin-LBackbone=Swin-L, Evaluation Protocol=Fine-tuned2022.02 | 47.8 | 81.6 | 89.7 | 75 | |
| Wukong_ViT-BBackbone=ViT-B, Evaluation Protocol=Fine-tuned2022.02 | 47.1 | 80.1 | 88.9 | 73.3 | |
| FILIP_Swin-LBackbone=Swin-L, Evaluation Protocol=Fine-tuned2022.02 | 45.2 | 77.9 | 87 | 71.9 | |
| FILIP_ViT-BBackbone=ViT-B, Evaluation Protocol=Fine-tuned2022.02 | 20.2 | 47.5 | 60.3 | 47.1 |