Text-to-image synthesis on MSCOCO (test)
10.5FIDOFA
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| OFAModel size=Large2022.02 | 10.5 | 34.4 | 31.1 | |
| GLIDEparameters=3.5B2022.02 | 12.2 | — | — | |
| NÜWA2022.02 | 12.9 | 34.3 | 27.2 | |
| VQ-Diffusionvariant=Full2021.11 | 13.86 | — | — | |
| VQ-Diffusionvariant=Base2021.11 | 19.75 | — | — | |
| DF-GAN2021.11 | 21.42 | — | — | |
| CogView2021.11 | 27.1 | — | — | |
| CogView2022.02 | 27.1 | 33.3 | 18.2 | |
| DALL-E2021.11 | 27.5 | — | — | |
| DALLE2022.02 | 27.5 | — | 17.9 | |
| DAE-GAN2021.11 | 28.12 | — | — | |
| Unifying2022.02 | 29.9 | 30.9 | — | |
| VQ-Diffusionvariant=Small2021.11 | 30.17 | — | — | |
| SEGAN2021.11 | 32.28 | — | — | |
| DM-GAN2021.11 | 32.64 | — | — | |
| AttnGAN2021.11 | 35.49 | — | — | |
| StackGAN2021.11 | 74.05 | — | — | |
| StackGAN++2021.11 | 81.59 | — | — |