cGANs with Projection Discriminator

ICLR 2018  ·  Takeru Miyato, Masanori Koyama ·

We propose a novel, projection based way to incorporate the conditional information into the discriminator of GANs that respects the role of the conditional information in the underlining probabilistic model. This approach is in contrast with most frameworks of conditional GANs used in application today, which use the conditional information by concatenating the (embedded) conditional vector to the feature vectors. With this modification, we were able to significantly improve the quality of the class conditional image generation on ILSVRC2012 (ImageNet) 1000-class image dataset from the current state-of-the-art result, and we achieved this with a single pair of a discriminator and a generator. We were also able to extend the application to super-resolution and succeeded in producing highly discriminative super-resolution images. This new structure also enabled high quality category transformation based on parametric functional transformation of conditional batch normalization layers in the generator.

PDF Abstract ICLR 2018 PDF ICLR 2018 Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Conditional Image Generation CIFAR-10 Projection Discriminator Inception score 8.62 # 10
FID 17.5 # 12
Conditional Image Generation ImageNet 128x128 Projection Discriminator FID 27.62 # 15
Inception score 36.8 # 14