Diffusion-GAN: Training GANs with Diffusion

Open paper
Benchmark Model Rank Results
image-generation-on-afhq-catDiffusion InsGen#5FID: 2.40
image-generation-on-afhq-dogDiffusion InsGen#3FID: 4.83
image-generation-on-afhq-wildDiffusion InsGen#3FID: 1.51
image-generation-on-celeba-64x64Diffusion StyleGAN2#4FID: 1.69
image-generation-on-ffhq-1024-x-1024Diffusion StyleGAN2#6FID: 2.83
image-generation-on-lsun-bedroom-256-x-256Diffusion ProjectedGAN#1FID: 1.43
image-generation-on-lsun-bedroom-256-x-256Diffusion StyleGAN2#6FID: 3.65
image-generation-on-lsun-bedroom-256-x-256Diffusion ProjectedGAN (DINOv2)#19FD: 547.61Precision: 0.79Recall: 0.28
image-generation-on-lsun-churches-256-x-256Diffusion ProjectedGAN#3FID: 1.85
image-generation-on-lsun-churches-256-x-256Diffusion StyleGAN2#6FID: 3.17
image-generation-on-stl-10Diffusion ProjectedGAN#1FID: 6.91
image-generation-on-stl-10Diffusion StyleGAN2#4FID: 11.53

Rank counts only results with a code link.