Image Synthesis on Cityscapes
3.41FIDProjected GAN
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| Projected GANResolution=256x2562021.11 | 3.41 | — | — | — | — | |
| Projected GANResolution=256x256, Training Point=First point surpassing SOTA2021.11 | 4.6 | — | — | — | — | |
| GANsformersResolution=256x2562021.11 | 5.23 | — | — | — | — | |
| GANformer_dAttention=Duplex2021.03 | 5.76 | — | 1.69 | 48.06 | 33.65 | |
| StyleGAN2-ADAResolution=256x2562021.11 | 8.35 | — | — | — | — | |
| StyleGAN22021.03 | 8.35 | — | 1.7 | 59.35 | 27.82 | |
| FastGANResolution=256x2562021.11 | 8.78 | — | — | — | — | |
| GAN2021.03 | 11.57 | — | 1.63 | 61.09 | 15.3 | |
| SAGANResolution=256x2562021.11 | 12.81 | — | — | — | — | |
| SAGAN2021.03 | 12.81 | — | 1.68 | 43.48 | 7.97 | |
| GANformer_sAttention=Simplex2021.03 | 14.23 | — | 1.67 | 64.12 | 2.03 | |
| k-GAN2021.03 | 51.08 | — | 1.66 | 18.8 | 1.73 |