Papers › Selective Frequency Network for Image Restoration
Selective Frequency Network for Image Restoration
Yuning Cui, Yi Tao, Zhenshan Bing, Wenqi Ren, Xinwei Gao, Xiaochun Cao, Kai Huang, Alois Knoll
Image restoration aims to reconstruct the latent sharp image from its corrupted counterpart. Besides dealing with this long-standing task in the spatial domain, a few approaches seek solutions in the frequency domain in consideration of the large discrepancy between spectra of sharp/degraded image pairs. However, these works commonly utilize transformation tools, e.g., wavelet transform, to split features into several frequency parts, which is not flexible enough to select the most informative frequency component to recover. In this paper, we exploit a multi-branch and content-aware module to decompose features into separate frequency subbands dynamically and locally, and then accentuate the useful ones via channel-wise attention weights. In addition, to handle large-scale degradation blurs, we propose an extremely simple decoupling and modulation module to enlarge the receptive field via global and window-based average pooling. Integrating two developed modules into a U-Net backbone, the proposed Selective Frequency Network (SFNet) performs favorably against state-of-the-art algorithms on five image restoration tasks, including single-image defocus deblurring, image dehazing, image motion deblurring, image desnowing, and image deraining.
Code
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
Results from the paper archive 2025-07-28
| Task | Dataset | Model | Metric | Value | Rank at snapshot | Leaderboard | Report |
|---|---|---|---|---|---|---|---|
| Deblurring | GoPro | SFNet | PSNR | 33.27 | #23 of 56 | Archive leaderboard | report |
| Deblurring | GoPro | SFNet | SSIM | 0.963 | #23 of 56 | Archive leaderboard | report |
| Deblurring | RSBlur | SFNet | Average PSNR | 34.35 | #3 of 12 | Archive leaderboard | report |
| Image Dehazing | SOTS Indoor | SFNet | PSNR | 41.24 | #10 of 34 | Archive leaderboard | report |
| Image Dehazing | SOTS Indoor | SFNet | SSIM | 0.996 | #10 of 34 | Archive leaderboard | report |
| Image Dehazing | SOTS Outdoor | SFNet | PSNR | 40.05 | #3 of 31 | Archive leaderboard | report |
| Image Dehazing | SOTS Outdoor | SFNet | SSIM | 0.996 | #3 of 31 | Archive leaderboard | report |
| Single Image Deraining | Rain100L | SFNet | PSNR | 38.21 | #9 of 19 | Archive leaderboard | report |
| Single Image Deraining | Rain100L | SFNet | SSIM | 0.974 | #9 of 19 | 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
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