DCI
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Text-to-Image Retrieval | DCI | R@179.7 | 106 | |
| Image-to-Text Retrieval | DCI | R@177.9 | 100 | |
| Text-to-Image Generation | DCI | FID22.1 | 26 | |
| Image-Text Retrieval | DCI long-text | Top-1 Accuracy63.4 | 22 | |
| Image-to-Text Retrieval | DCI (test) | R@160.76 | 12 | |
| Crossmodal retrieval | DCI | R@175.9 | 12 | |
| Text-to-Image Retrieval | DCI-CN | Recall@165.1 | 11 | |
| Super-Resolution | DCI x4 (test) | PSNR31.73 | 7 | |
| Super-Resolution | DCI x3 (test) | PSNR33.42 | 7 | |
| Super-Resolution | DCI x2 (test) | PSNR37.55 | 7 | |
| Image Inversion and Reconstruction | DCI (Densely Captioned Images) (first 1K images) | LPIPS0.1579 | 7 | |
| Text-to-Image Retrieval | DCI (test) | R@159.73 | 6 |
Showing 12 of 12 rows