Papers › HINet: Half Instance Normalization Network for Image Restoration

HINet: Half Instance Normalization Network for Image Restoration

13 May 2021arXiv:2105.06086archive 2025-07-28

Liangyu Chen, Xin Lu, Jie Zhang, Xiaojie Chu, Chengpeng Chen

In this paper, we explore the role of Instance Normalization in low-level vision tasks. Specifically, we present a novel block: Half Instance Normalization Block (HIN Block), to boost the performance of image restoration networks. Based on HIN Block, we design a simple and powerful multi-stage network named HINet, which consists of two subnetworks. With the help of HIN Block, HINet surpasses the state-of-the-art (SOTA) on various image restoration tasks. For image denoising, we exceed it 0.11dB and 0.28 dB in PSNR on SIDD dataset, with only 7.5% and 30% of its multiplier-accumulator operations (MACs), 6.8 times and 2.9 times speedup respectively. For image deblurring, we get comparable performance with 22.5% of its MACs and 3.3 times speedup on REDS and GoPro datasets. For image deraining, we exceed it by 0.3 dB in PSNR on the average result of multiple datasets with 1.4 times speedup. With HINet, we won 1st place on the NTIRE 2021 Image Deblurring Challenge - Track2. JPEG Artifacts, with a PSNR of 29.70. The code is available at https://github.com/megvii-model/HINet.

PaperPDFCode

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

Code

megvii-model/HINet officialmentioned in papermentioned on GitHubpytorchNOASSERTION report
akalia77/hinet mentioned on GitHubpytorchNOASSERTION 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

DeblurringDenoisingImage DeblurringImage DenoisingImage RestorationRain RemovalSingle Image DerainingSpectral Reconstruction

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Deblurring GoPro HINet PSNR 32.71 #31 of 56 Archive leaderboard report
Image Denoising SIDD HINet PSNR (sRGB) 39.99 #7 of 22 Archive leaderboard report
Image Denoising SIDD HINet SSIM (sRGB) 0.958 #7 of 22 Archive leaderboard report
Single Image Deraining Rain100H HINet PSNR 30.65 #8 of 19 Archive leaderboard report
Single Image Deraining Rain100H HINet SSIM 0.894 #8 of 19 Archive leaderboard report
Single Image Deraining Rain100L HINet PSNR 37.28 #11 of 19 Archive leaderboard report
Single Image Deraining Rain100L HINet SSIM 0.97 #11 of 19 Archive leaderboard report
Single Image Deraining Test100 HINet PSNR 30.29 #5 of 12 Archive leaderboard report
Single Image Deraining Test100 HINet SSIM 0.906 #5 of 12 Archive leaderboard report
Single Image Deraining Test1200 HINet PSNR 33.05 #6 of 14 Archive leaderboard report
Single Image Deraining Test1200 HINet SSIM 0.919 #6 of 14 Archive leaderboard report
Single Image Deraining Test2800 HINet PSNR 33.91 #3 of 12 Archive leaderboard report
Single Image Deraining Test2800 HINet SSIM 0.941 #3 of 12 Archive leaderboard report
Spectral Reconstruction ARAD-1K HINet MRAE 0.2032 #6 of 11 Archive leaderboard report
Spectral Reconstruction ARAD-1K HINet PSNR 32.51 #6 of 11 Archive leaderboard report
Spectral Reconstruction ARAD-1K HINet RMSE 0.0303 #6 of 11 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

Instance 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