FreGAN: Exploiting Frequency Components for Training GANs under Limited Data

TL;DR

FreGAN enhances GAN training with limited data by leveraging frequency information, especially high-frequency signals, to improve generation quality and reduce overfitting, outperforming existing methods in low-data scenarios.

Contribution

FreGAN introduces a frequency-aware training approach that incorporates frequency signals of real and generated images, including a self-supervised constraint, to improve GAN performance with limited data.

Findings

Outperforms existing methods in low-data regimes (<100 samples)

Effectively utilizes high-frequency information for better detail preservation

Can be integrated with other regularization and attention models

Abstract

Training GANs under limited data often leads to discriminator overfitting and memorization issues, causing divergent training. Existing approaches mitigate the overfitting by employing data augmentations, model regularization, or attention mechanisms. However, they ignore the frequency bias of GANs and take poor consideration towards frequency information, especially high-frequency signals that contain rich details. To fully utilize the frequency information of limited data, this paper proposes FreGAN, which raises the model's frequency awareness and draws more attention to producing high-frequency signals, facilitating high-quality generation. In addition to exploiting both real and generated images' frequency information, we also involve the frequency signals of real images as a self-supervised constraint, which alleviates the GAN disequilibrium and encourages the generator to synthesize adequate rather than arbitrary frequency signals. Extensive results demonstrate the superiority and effectiveness of our FreGAN in ameliorating generation quality in the low-data regime (especially when training data is less than 100). Besides, FreGAN can be seamlessly applied to existing regularization and attention mechanism models to further boost the performance.