HP-GAN: Harnessing pretrained networks for GAN improvement with FakeTwins and discriminator consistency

Open paper
Benchmark Model Rank Results
image-generation-on-afhq-dogHP-GAN#1FID: 3.63
image-generation-on-lsun-bedroom-256-x-256HP-GAN#1FID: 1.19
image-generation-on-lsun-churches-256-x-256HP-GAN#1FID: 1.44
image-generation-on-ffhq-256-x-256HP-GAN#2FID: 1.69

Rank counts only results with a code link.