StyleGAN-XL (Exposing)

Reported on 6 benchmarks across 1 task · 1 paper

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Medical6 results

Image GenerationonFFHQ 256 x 256
Coverage· 2023-06-07
0.63
best: 0.74 (LDM (Exposing))
Exposing flaws of generative model evaluation metrics and their unfair treatment of diffusion models arXiv:2306.04675
Image GenerationonFFHQ 256 x 256
Density· 2023-06-07
0.68
best: 1.04 (Efficient-vdVAE (Exposing))
Exposing flaws of generative model evaluation metrics and their unfair treatment of diffusion models arXiv:2306.04675
Image GenerationonFFHQ 256 x 256
FD· 2023-06-07
240.07
best: 592.26 (Projected-GAN (DINOv2))
Exposing flaws of generative model evaluation metrics and their unfair treatment of diffusion models arXiv:2306.04675
Image GenerationonFFHQ 256 x 256
FID· 2023-06-07
2.26
best: 1.68 (StyleSAN-XL)
Exposing flaws of generative model evaluation metrics and their unfair treatment of diffusion models arXiv:2306.04675
Image GenerationonFFHQ 256 x 256
Precision· 2023-06-07
0.77
best: 0.92 (LDM (DINOv2))
Exposing flaws of generative model evaluation metrics and their unfair treatment of diffusion models arXiv:2306.04675
Image GenerationonFFHQ 256 x 256
Recall· 2023-06-07
0.43
best: 0.76 (LDM (DINOv2))
Exposing flaws of generative model evaluation metrics and their unfair treatment of diffusion models arXiv:2306.04675