cGANs with Projection Discriminator

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... (read more)

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 # 5
FID 17.5 # 6
Conditional Image Generation ImageNet 128x128 Projection Discriminator FID 27.62 # 10
Inception score 36.8 # 9

Methods used in the Paper