Papers › Exploring The Effect of High-frequency Components in GANs Training

Exploring The Effect of High-frequency Components in GANs Training

20 Mar 2021arXiv:2103.11093archive 2025-07-28

Ziqiang Li, Pengfei Xia, Xue Rui, Bin Li

Generative Adversarial Networks (GANs) have the ability to generate images that are visually indistinguishable from real images. However, recent studies have revealed that generated and real images share significant differences in the frequency domain. In this paper, we explore the effect of high-frequency components in GANs training. According to our observation, during the training of most GANs, severe high-frequency differences make the discriminator focus on high-frequency components excessively, which hinders the generator from fitting the low-frequency components that are important for learning images' content. Then, we propose two simple yet effective frequency operations for eliminating the side effects caused by high-frequency differences in GANs training: High-Frequency Confusion (HFC) and High-Frequency Filter (HFF). The proposed operations are general and can be applied to most existing GANs with a fraction of the cost. The advanced performance of the proposed operations is verified on multiple loss functions, network architectures, and datasets. Specifically, the proposed HFF achieves significant improvements of 42.5% FID on CelebA (128*128) unconditional generation based on SNGAN, 30.2% FID on CelebA unconditional generation based on SSGAN, and 69.3% FID on CelebA unconditional generation based on InfoMAXGAN.

PaperPDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

kwotsin/mimicry officialmentioned in paperpytorch report
iceli1007/HFC-and-HFF mentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Vocal Bursts Intensity Prediction

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

GAN Hinge LossSNGANSpectral Normalization

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections