Large Scale GAN Training for High Fidelity Natural Image Synthesis

Open paper
Benchmark Model Rank Results
conditional-image-generation-on-artbench-10BigGAN + DiffAug#3FID: 4.055
conditional-image-generation-on-cifar-10BigGAN#13FID: 14.73Inception score: 9.22
conditional-image-generation-on-imagenetBigGAN-deep#7FID: 5.7Inception score: 124.5
conditional-image-generation-on-imagenetBigGAN#13FID: 8.7Inception score: 98.8
image-generation-on-cifar-10BigGAN#61IS: 9.22
image-generation-on-imagenet-128x128BigGAN-deep#10FID: 5.7IS: 124.5
image-generation-on-imagenet-128x128BigGAN#13FID: 8.7IS: 98.8
image-generation-on-imagenet-256x256BigGAN-deep#84FID: 8.1

Rank counts only results with a code link.