Text-to-Image Retrieval on Flickr8K-CN
70.1R@1Wukong_ViT-L
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Wukong_ViT-LBackbone=ViT-L, Evaluation Protocol=Fine-tuned2022.02 | 70.1 | 91.9 | 96.4 | 89.7 | |
| FILIP_ViT-LBackbone=ViT-L, Evaluation Protocol=Fine-tuned2022.02 | 68.5 | 90.5 | 95.2 | 88 | |
| CLIP_ViT-LBackbone=ViT-L, Evaluation Protocol=Fine-tuned2022.02 | 67.4 | 91 | 95.7 | 88.6 | |
| FILIP_Swin-LBackbone=Swin-L, Evaluation Protocol=Fine-tuned2022.02 | 61.5 | 86.5 | 93 | 85.1 | |
| CLIP_ViT-BBackbone=ViT-B, Evaluation Protocol=Fine-tuned2022.02 | 61.2 | 86.8 | 93.2 | 85.3 | |
| CLIP_Swin-LBackbone=Swin-L, Evaluation Protocol=Fine-tuned2022.02 | 59.3 | 86 | 92.9 | 84.8 | |
| Wukong_ViT-BBackbone=ViT-B, Evaluation Protocol=Fine-tuned2022.02 | 58.4 | 85.4 | 92 | 82.6 | |
| Wukong_Swin-LBackbone=Swin-L, Evaluation Protocol=Fine-tuned2022.02 | 57.9 | 85.1 | 92.6 | 83.6 | |
| FILIPBackbone=ViT-L, Zero-shot=true2022.02 | 53.5 | 79.3 | 87.9 | 79.8 | |
| CLIPBackbone=ViT-L, Zero-shot=true2022.02 | 50.5 | 77 | 85.7 | 77.2 | |
| FILIP_ViT-BBackbone=ViT-B, Evaluation Protocol=Fine-tuned2022.02 | 46.4 | 77 | 86.8 | 72.4 | |
| WukongBackbone=ViT-L, Zero-shot=true2022.02 | 46 | 74.5 | 84.5 | 74.4 | |
| CLIPBackbone=ViT-B, Zero-shot=true2022.02 | 44.2 | 71.2 | 82 | 72.7 | |
| WukongBackbone=ViT-B, Zero-shot=true2022.02 | 43.2 | 71.3 | 81.3 | 70.6 | |
| FILIPBackbone=Swin-L, Zero-shot=true2022.02 | 41.2 | 68.5 | 79.1 | 67.7 | |
| CLIPBackbone=Swin-L, Zero-shot=true2022.02 | 38.6 | 67 | 78.2 | 69.2 | |
| WukongBackbone=Swin-L, Zero-shot=true2022.02 | 36.6 | 64.8 | 76.2 | 65.2 | |
| FILIPBackbone=ViT-B, Zero-shot=true2022.02 | 24 | 50 | 62.4 | 52.5 | |
| BriVLBackbone=N/A, Zero-shot=true2022.02 | 8 | 20.7 | 29.5 | 23.9 |