Papers › MSSNet: Multi-Scale-Stage Network for Single Image Deblurring

MSSNet: Multi-Scale-Stage Network for Single Image Deblurring

19 Feb 2022arXiv:2202.09652archive 2025-07-28

Kiyeon Kim, Seungyong Lee, Sunghyun Cho

Most of traditional single image deblurring methods before deep learning adopt a coarse-to-fine scheme that estimates a sharp image at a coarse scale and progressively refines it at finer scales. While this scheme has also been adopted to several deep learning-based approaches, recently a number of single-scale approaches have been introduced showing superior performance to previous coarse-to-fine approaches both in quality and computation time. In this paper, we revisit the coarse-to-fine scheme, and analyze defects of previous coarse-to-fine approaches that degrade their performance. Based on the analysis, we propose Multi-Scale-Stage Network (MSSNet), a novel deep learning-based approach to single image deblurring that adopts our remedies to the defects. Specifically, MSSNet adopts three novel technical components: stage configuration reflecting blur scales, an inter-scale information propagation scheme, and a pixel-shuffle-based multi-scale scheme. Our experiments show that MSSNet achieves the state-of-the-art performance in terms of quality, network size, and computation time.

PaperPDFCode

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

Code

kky7/MSSNet officialpytorch 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

DeblurringDeep LearningImage DeblurringSingle Image Deblurring

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Deblurring GoPro MSSNet-large PSNR 33.39 #20 of 56 Archive leaderboard report
Deblurring GoPro MSSNet-large SSIM 0.964 #20 of 56 Archive leaderboard report
Deblurring GoPro MSSNet PSNR 33.01 #27 of 56 Archive leaderboard report
Deblurring GoPro MSSNet SSIM 0.961 #27 of 56 Archive leaderboard report
Deblurring GoPro MSSNet-small PSNR 32.02 #39 of 56 Archive leaderboard report
Deblurring GoPro MSSNet-small SSIM 0.953 #39 of 56 Archive leaderboard report
Deblurring RealBlur-J MSSNet PSNR (sRGB) 32.1 #12 of 17 Archive leaderboard report
Deblurring RealBlur-J MSSNet Params(M) 15.6 #12 of 17 Archive leaderboard report
Deblurring RealBlur-J MSSNet SSIM (sRGB) 0.928 #12 of 17 Archive leaderboard report
Deblurring RealBlur-J (trained on GoPro) MSSNet PSNR (sRGB) 28.79 #9 of 15 Archive leaderboard report
Deblurring RealBlur-J (trained on GoPro) MSSNet SSIM (sRGB) 0.879 #9 of 15 Archive leaderboard report
Deblurring RealBlur-R MSSNet PSNR (sRGB) 39.76 #11 of 17 Archive leaderboard report
Deblurring RealBlur-R MSSNet Params 15.59 #11 of 17 Archive leaderboard report
Deblurring RealBlur-R MSSNet SSIM (sRGB) 0.972 #11 of 17 Archive leaderboard report
Deblurring RealBlur-R (trained on GoPro) MSSNet PSNR (sRGB) 35.93 #9 of 19 Archive leaderboard report
Deblurring RealBlur-R (trained on GoPro) MSSNet SSIM (sRGB) 0.953 #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.

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