Simple yet Effective Way for Improving the Performance of GAN

TL;DR

This paper introduces a simple cascading rejection module for GAN discriminators that extracts diverse features to improve the generator’s quality without increasing training complexity.

Contribution

It proposes a novel cascading rejection module that enhances discriminator robustness and GAN performance with minimal additional computational overhead.

Findings

Significant improvement in FID scores across multiple datasets.

Enhanced diversity and visual quality of generated images.

Method is easy to integrate into existing GAN frameworks.

Abstract

In adversarial learning, discriminator often fails to guide the generator successfully since it distinguishes between real and generated images using silly or non-robust features. To alleviate this problem, this brief presents a simple but effective way that improves the performance of generative adversarial network (GAN) without imposing the training overhead or modifying the network architectures of existing methods. The proposed method employs a novel cascading rejection (CR) module for discriminator, which extracts multiple non-overlapped features in an iterative manner using the vector rejection operation. Since the extracted diverse features prevent the discriminator from concentrating on non-meaningful features, the discriminator can guide the generator effectively to produce the images that are more similar to the real images. In addition, since the proposed CR module requires only a few simple vector operations, it can be readily applied to existing frameworks with marginal training overheads. Quantitative evaluations on various datasets including CIFAR-10, CelebA, CelebA-HQ, LSUN, and tiny-ImageNet confirm that the proposed method significantly improves the performance of GAN and conditional GAN in terms of Frechet inception distance (FID) indicating the diversity and visual appearance of the generated images.