Omni-GAN: On the Secrets of cGANs and Beyond

Abstract

It has been an important problem to design a proper discriminator for conditional generative adversarial networks (cGANs). In this paper, we investigate two popular choices, the projection-based and classification-based discriminators, and reveal that both of them suffer some kind of drawbacks that affect the learning ability of cGANs. Then, we present our solution that trains a powerful discriminator and avoids over-fitting with regularization. In addition, we unify multiple targets (class, domain, reality, etc.) into one loss function to enable a wider range of applications. Our algorithm, named Omni-GAN, achieves competitive performance on a few popular benchmarks. More importantly, Omni-GAN enjoys both high generation quality and low risks in mode collapse, offering new possibilities for optimizing cGANs. Code is available 1 .

Conditional GANs with Auxiliary Discriminative Classifier

“…To implement the conditional generator, common technique nowadays injects the conditional information via conditional batch normalization [4]. To train the conditional generator, a lot of researches focus on effectively injecting the conditional information into the discriminator [21,19,34,13,9,33]. Among them, the auxiliary classifier generative adversarial networks (AC-GAN) [22] have been widely used due to its simplicity and extensibility.…”