Papers › DeblurGAN-v2: Deblurring (Orders-of-Magnitude) Faster and Better

DeblurGAN-v2: Deblurring (Orders-of-Magnitude) Faster and Better

10 Aug 2019ICCV 2019 10arXiv:1908.03826archive 2025-07-28

Orest Kupyn, Tetiana Martyniuk, Junru Wu, Zhangyang Wang

We present a new end-to-end generative adversarial network (GAN) for single image motion deblurring, named DeblurGAN-v2, which considerably boosts state-of-the-art deblurring efficiency, quality, and flexibility. DeblurGAN-v2 is based on a relativistic conditional GAN with a double-scale discriminator. For the first time, we introduce the Feature Pyramid Network into deblurring, as a core building block in the generator of DeblurGAN-v2. It can flexibly work with a wide range of backbones, to navigate the balance between performance and efficiency. The plug-in of sophisticated backbones (e.g., Inception-ResNet-v2) can lead to solid state-of-the-art deblurring. Meanwhile, with light-weight backbones (e.g., MobileNet and its variants), DeblurGAN-v2 reaches 10-100 times faster than the nearest competitors, while maintaining close to state-of-the-art results, implying the option of real-time video deblurring. We demonstrate that DeblurGAN-v2 obtains very competitive performance on several popular benchmarks, in terms of deblurring quality (both objective and subjective), as well as efficiency. Besides, we show the architecture to be effective for general image restoration tasks too. Our codes, models and data are available at: https://github.com/KupynOrest/DeblurGANv2

PaperPDFConference PDFCode

In Syntology View this paper on Syntology, its page in Syntology's graph. That page lists the repositories linked to the paper, the abstract and the calls for agents.

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

For agents, on Syntology's MCP service (how to connect):

Code

KupynOrest/DeblurGANv2 officialmentioned in papermentioned on GitHubpytorchNOASSERTIONSyntology: not harvested report
HDCVLab/MC-Blur-Dataset mentioned on GitHubpytorchSyntology: not harvested report
TAMU-VITA/DeblurGANv2 mentioned on GitHubpytorchNOASSERTIONSyntology: not harvested report
kritiksoman/GIMP-ML mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to RIFE: Real-Time Intermediate Flow Estimation for Video Frame Interpolation (arXiv:2011.06294; that paper's own run record: 19 ran (of which 5 constructed an object rather than computing a result; 9 with no instrument failure: 1 honoured, 0 violated, 8 with no contract checked; 10 where Syntology's instrument failed) · 7 unverified) and Vision Transformers for Dense Prediction (arXiv:2103.13413; that paper's own run record: 65 ran (of which 19 constructed an object rather than computing a result; 32 with no instrument failure: 0 honoured, 0 violated, 32 with no contract checked; 33 where Syntology's instrument failed) · 51 unverified); nothing checks that they implement this paper's method. report
vcarehuman/YoloPose mentioned on GitHubpytorchNOASSERTIONSyntology: not harvested report
vita-group/deblurganv2 mentioned on GitHubpytorchNOASSERTIONSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to DMPlug: A Plug-in Method for Solving Inverse Problems with Diffusion Models (arXiv:2405.16749; that paper's own run record: 41 ran (of which 6 constructed an object rather than computing a result; 18 with no instrument failure: 5 honoured, 0 violated, 13 with no contract checked; 23 where Syntology's instrument failed) · 19 unverified); nothing checks that they implement this paper's method. 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.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

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

Tasks

Blind Face RestorationDeblurringImage DeblurringImage RestorationNavigateSingle-Image Blind DeblurringVideo Deblurring

1 archive task tag without a task page not shown.

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Blind Face Restoration CelebA-Test DeblurGANv2* Deg. 39.64 #3 of 15 Archive leaderboard report
Blind Face Restoration CelebA-Test DeblurGANv2* FID 52.69 #3 of 15 Archive leaderboard report
Blind Face Restoration CelebA-Test DeblurGANv2* LPIPS 40.01 #3 of 15 Archive leaderboard report
Blind Face Restoration CelebA-Test DeblurGANv2* NIQE 4.917 #3 of 15 Archive leaderboard report
Blind Face Restoration CelebA-Test DeblurGANv2* PSNR 25.91 #3 of 15 Archive leaderboard report
Blind Face Restoration CelebA-Test DeblurGANv2* SSIM 0.6952 #3 of 15 Archive leaderboard report
Deblurring BASED DeblurGAN Inception ERQAv2.0 0.74297 #2 of 4 Archive leaderboard report
Deblurring BASED DeblurGAN Inception LPIPS 0.08867 #2 of 4 Archive leaderboard report
Deblurring BASED DeblurGAN Inception PSNR 31.17171 #2 of 4 Archive leaderboard report
Deblurring BASED DeblurGAN Inception SSIM 0.94301 #2 of 4 Archive leaderboard report
Deblurring BASED DeblurGAN Inception VMAF 66.91781 #2 of 4 Archive leaderboard report
Deblurring GoPro DeblurGAN-v2 PSNR 29.55 #50 of 56 Archive leaderboard report
Deblurring GoPro DeblurGAN-v2 SSIM 0.934 #50 of 56 Archive leaderboard report
Deblurring GoPro DeblurGANv2-MobileNet PSNR 28.17 #54 of 56 Archive leaderboard report
Deblurring GoPro DeblurGANv2-MobileNet SSIM 0.925 #54 of 56 Archive leaderboard report
Deblurring GoPro DeblurGANv2-MobileNet-DSC PSNR 28.03 #55 of 56 Archive leaderboard report
Deblurring GoPro DeblurGANv2-MobileNet-DSC SSIM 0.922 #55 of 56 Archive leaderboard report
Deblurring MSU BASED DeblurGAN Inception Subjective 1.0375 #4 of 11 Archive leaderboard report
Deblurring RealBlur-J DeblurGAN-v2 PSNR (sRGB) 29.69 #17 of 17 Archive leaderboard report
Deblurring RealBlur-J DeblurGAN-v2 Params(M) 5.08 #17 of 17 Archive leaderboard report
Deblurring RealBlur-J DeblurGAN-v2 SSIM (sRGB) 0.870 #17 of 17 Archive leaderboard report
Deblurring RealBlur-J (trained on GoPro) DeblurGAN-v2 PSNR (sRGB) 28.70 #11 of 15 Archive leaderboard report
Deblurring RealBlur-J (trained on GoPro) DeblurGAN-v2 SSIM (sRGB) 0.866 #11 of 15 Archive leaderboard report
Deblurring RealBlur-R DeblurGAN-v2 PSNR (sRGB) 36.44 #16 of 17 Archive leaderboard report
Deblurring RealBlur-R DeblurGAN-v2 SSIM (sRGB) 0.935 #16 of 17 Archive leaderboard report
Deblurring RealBlur-R (trained on GoPro) DeblurGAN-v2 SSIM (sRGB) 0.944 #16 of 19 Archive leaderboard report
Image Deblurring GoPro DeblurGAN-v2 PSNR 29.55 #49 of 55 Archive leaderboard report
Image Deblurring GoPro DeblurGAN-v2 Params (M) 5.08 #49 of 55 Archive leaderboard report
Image Deblurring GoPro DeblurGAN-v2 SSIM 0.925 #49 of 55 Archive leaderboard report
Image Deblurring GoPro DeblurGANv2-MobileNet PSNR 28.17 #52 of 55 Archive leaderboard report
Image Deblurring GoPro DeblurGANv2-MobileNet-DSC PSNR 28.03 #53 of 55 Archive leaderboard report
Image Deblurring GoPro DeblurGANv2-MobileNet-DSC SSIM 0.922 #53 of 55 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

Convolution

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