Papers › Focal Frequency Loss for Image Reconstruction and Synthesis

Focal Frequency Loss for Image Reconstruction and Synthesis

23 Dec 2020ICCV 2021 10arXiv:2012.12821archive 2025-07-28

Liming Jiang, Bo Dai, Wayne Wu, Chen Change Loy

Image reconstruction and synthesis have witnessed remarkable progress thanks to the development of generative models. Nonetheless, gaps could still exist between the real and generated images, especially in the frequency domain. In this study, we show that narrowing gaps in the frequency domain can ameliorate image reconstruction and synthesis quality further. We propose a novel focal frequency loss, which allows a model to adaptively focus on frequency components that are hard to synthesize by down-weighting the easy ones. This objective function is complementary to existing spatial losses, offering great impedance against the loss of important frequency information due to the inherent bias of neural networks. We demonstrate the versatility and effectiveness of focal frequency loss to improve popular models, such as VAE, pix2pix, and SPADE, in both perceptual quality and quantitative performance. We further show its potential on StyleGAN2.

PaperPDFConference PDFCode

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

Code

EndlessSora/focal-frequency-loss officialmentioned in papermentioned 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

Image GenerationImage ReconstructionImage-to-Image Translation

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Image-to-Image Translation Cityscapes Labels-to-Photo SPADE + FFL FID 59.5 #6 of 21 Archive leaderboard report
Image-to-Image Translation Cityscapes Labels-to-Photo SPADE + FFL Per-pixel Accuracy 82.5% #6 of 21 Archive leaderboard report
Image-to-Image Translation Cityscapes Labels-to-Photo SPADE + FFL mIoU 64.2 #6 of 21 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

ConvolutionPath Length RegularizationR1 RegularizationSPADEStyleGAN2Weight Demodulation

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