Papers › Zero-Shot Image Restoration Using Denoising Diffusion Null-Space Model

Zero-Shot Image Restoration Using Denoising Diffusion Null-Space Model

1 Dec 2022arXiv:2212.00490archive 2025-07-28

Yinhuai Wang, Jiwen Yu, Jian Zhang

Most existing Image Restoration (IR) models are task-specific, which can not be generalized to different degradation operators. In this work, we propose the Denoising Diffusion Null-Space Model (DDNM), a novel zero-shot framework for arbitrary linear IR problems, including but not limited to image super-resolution, colorization, inpainting, compressed sensing, and deblurring. DDNM only needs a pre-trained off-the-shelf diffusion model as the generative prior, without any extra training or network modifications. By refining only the null-space contents during the reverse diffusion process, we can yield diverse results satisfying both data consistency and realness. We further propose an enhanced and robust version, dubbed DDNM+, to support noisy restoration and improve restoration quality for hard tasks. Our experiments on several IR tasks reveal that DDNM outperforms other state-of-the-art zero-shot IR methods. We also demonstrate that DDNM+ can solve complex real-world applications, e.g., old photo restoration.

PaperPDFCodeCode Syntology ran

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also 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

No sample from the archive's official repository (wyhuai/ddnm) is linked to this paper; the 23 samples below come from 1 community repository. Syntology harvested code from wyhuai/ddnm for another paper.

wyhuai/ddnm: Syntology's graph links these samples to Unlimited-Size Diffusion Restoration (arXiv:2303.00354; that paper's own run record: 20 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 1 honoured, 1 violated, 2 with no contract checked; 16 where Syntology's instrument failed) · 2 unverified); nothing checks that they implement this paper's method.

Syntology 23 code samples harvested from 1 repository linked to this paper: 16 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 1 honoured, 0 violated, 3 with no contract checked; 12 where Syntology's instrument failed) · 7 unverified. Where Syntology's instrument failed: 7 where our contract draft was wrong; 5 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: community (archive-listed): 23 samples from 1 repository, 16 ran (4 with no instrument failure; 12 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

No sample is linked to this paper from 2 of its repositories that Syntology harvested for other papers:

Levels written between 2026-09-22 and 2026-09-26 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed. 4 of the 16 samples that ran have no recorded date.

wyhuai/ddnm officialmentioned in papermentioned on GitHubpytorchMITSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Unlimited-Size Diffusion Restoration (arXiv:2303.00354; that paper's own run record: 20 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 1 honoured, 1 violated, 2 with no contract checked; 16 where Syntology's instrument failed) · 2 unverified); nothing checks that they implement this paper's method. report
andreamazzitelli/ProjectNN mentioned on GitHubSyntology: not harvested report
ipc-lab/deepjscc-diffusion mentioned on GitHubpytorchMITSyntology: 23 samples, 16 ran (12 instrument-failed) · 7 unverified report
xypeng9903/k-diffusion-inverse-problems mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Diffusion Posterior Sampling for General Noisy Inverse Problems (arXiv:2209.14687; that paper's own run record: 6 ran (of which 4 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 1 violated, 4 with no contract checked; 1 where Syntology's instrument failed) · 14 unverified) and Denoising Diffusion Models for Plug-and-Play Image Restoration (arXiv:2305.08995; that paper's own run record: 15 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 3 honoured, 0 violated, 2 with no contract checked; 10 where Syntology's instrument failed) · 1 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

23 samples harvested: 16 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 1 honoured, 0 violated, 3 with no contract checked; 12 where Syntology's instrument failed) · 7 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

1ran · honoured contract
7ran · our draft was wrong
5ran; the contract check could not run
3ran
7unverified

Licence: 10 of the 23 samples are pointer only, meaning Syntology does not serve that copy's text. For 10 of them, this paper's copy is under a permissive licence and cleared, but the stored copy of the same code that Syntology serves is not; the licence label's tooltip gives Syntology's recorded reason. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from ipc-lab/deepjscc-diffusion. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

ipc-lab/deepjscc-diffusion community (archive-listed) 23 samples, 16 ran (12 instrument-failed) · 7 unverified
MeanUpsample guided_diffusion/diffusion.py ran; the contract check could not run (RAISES) · metamorphic tier: invariant fingerprinted MIT (permissive) · 12012b9130641107 · report fixture: signature x:Tensor/4/float, scale:int/0/int · returned Tensor · level recorded, date not recorded
Normalize in 154 papers guided_diffusion/models.py ran · our draft was wrong MIT (permissive) here; the stored copy is pointer only · c3a6b977022957cb · report fixture: signature in_channels:int/0/int · returned GroupNorm · level recorded, date not recorded
color2gray in 7 papers guided_diffusion/diffusion.py ran; the contract check could not run (RAISES) fingerprinted MIT (permissive) here; the stored copy is pointer only · af792c258aa85c57 · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-24
default_loader in 66 papers datasets/imagenet_subset.py ran; the contract check could not run (DEP_MISSING) MIT (permissive) here; the stored copy is pointer only · 7bf07d5f59aae36d · report fixture: signature path:str/0/str · returned Image · level recorded 2026-09-25
detect_mask sampler.py ran fingerprinted MIT (permissive) · ea9bfa7ba331cfcf · report fixture: signature y0:Tensor/4/float, thres:float/0/float · returned Tensor · level recorded 2026-09-25
get_gaussian_noisy_img guided_diffusion/diffusion.py ran fingerprinted MIT (permissive) here; the stored copy is pointer only · 17b8a580da43c920 · report fixture: signature img:Tensor/4/float, noise_level:float/0/float · returned Tensor · level recorded 2026-09-24
get_param_groups_and_shapes in 73 papers guided_diffusion/fp16_util.py ran · our draft was wrong MIT (permissive) here; the stored copy is pointer only · e41367ad14ff58fd · report fixture: signature named_model_params:list/1/tuple · returned list · level recorded 2026-09-24
get_timestep_embedding in 138 papers
census: 11 implementations from 15 papers, 5 distinct outputs across 2 buckets, one shared input per bucket this sample: bucket 1 of 2, arg 1: rank 1, kind int, dtype int64
guided_diffusion/models.py ran; the contract check could not run (RAISES) · metamorphic tier: invariant MIT (permissive) · cb49209c125de1b4 · report fixture: signature timesteps:Tensor/1/int, embedding_dim:int/0/int · returned Tensor · level recorded, date not recorded
make_master_params in 90 papers guided_diffusion/fp16_util.py ran · our draft was wrong MIT (permissive) here; the stored copy is pointer only · e20dd5102da3b050 · report fixture: signature param_groups_and_shapes:list/1/tuple · returned list · level recorded 2026-09-24
make_output_format in 88 papers guided_diffusion/logger.py ran · our draft was wrong MIT (permissive) · bcd8b4acab199405 · report fixture: signature format:str/0/str, ev_dir:str/0/str, log_suffix:str/0/str · returned HumanOutputFormat · level recorded 2026-09-24
masking_regularizer sampler.py ran MIT (permissive) · a59051041a1361a9 · report fixture: signature y0:Tensor/4/float, x0:Tensor/4/float, cond_kwargs:dict/0/? · returned Tensor · level recorded 2026-09-25
mpi_weighted_mean in 55 papers guided_diffusion/logger.py ran · our draft was wrong MIT (permissive) here; the stored copy is pointer only · e515a67f7f32e76d · report fixture: signature comm:MockComm/0/?, local_name2valcount:dict/0/? · returned dict · level recorded 2026-09-26
nonlinearity in 140 papers
census: 4 implementations from 11 papers, 1 distinct output on one shared input this sample: bucket 1 of 1, arg 1: rank 2, kind float, dtype float32
guided_diffusion/models.py ran · our draft was wrong fingerprinted MIT (permissive) · 3137073275f8c21a · report fixture: signature x:Tensor/2/float · returned Tensor · level recorded, date not recorded
pil_loader in 144 papers datasets/imagenet_subset.py ran · honoured contract MIT (permissive) · f321f54723433661 · report fixture: signature path:str/0/str · returned Image · level recorded 2026-09-22
unflatten_master_params in 90 papers guided_diffusion/fp16_util.py ran; the contract check could not run (RAISES) MIT (permissive) here; the stored copy is pointer only · 64fff1e30802b815 · report fixture: signature param_group:list/1/tuple, master_param:Tensor/1/float · returned tuple · level recorded 2026-09-24
zero_module in 184 papers guided_diffusion/nn.py ran · our draft was wrong MIT (permissive) · 129b804760b3115f · report fixture: signature module:callable/0/? · returned SimpleModule · level recorded 2026-09-22
accimage_loader in 47 papers datasets/imagenet_subset.py unverified MIT (permissive) here; the stored copy is pointer only · 404fb2b2daa1ae78 · report
avg_pool_nd in 83 papers guided_diffusion/nn.py unverified MIT (permissive) · ecd0fc28815b65ae · report
conv_nd in 94 papers guided_diffusion/nn.py unverified MIT (permissive) · fe4eb545bbb728e0 · report
create_classifier_and_diffusion in 11 papers guided_diffusion/script_util.py unverified MIT (permissive) · 5e5d99a81705a07f · report
create_model in 163 papers guided_diffusion/script_util.py unverified MIT (permissive) · 7e05e674886fa857 · report
create_model_and_diffusion in 18 papers guided_diffusion/script_util.py unverified MIT (permissive) · dd3e49ff96ee5e82 · report
profile in 89 papers guided_diffusion/logger.py unverified MIT (permissive) here; the stored copy is pointer only · 0c6607473a4c4c55 · report

Tasks

ColorizationDeblurringDenoisingImage ColorizationImage Compressed SensingImage DeblurringImage InpaintingImage RestorationImage Super-ResolutionSuper-Resolutioncompressed sensing

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Image Colorization CelebA DDRM Consistency 455.9 #1 of 3 Archive leaderboard report
Image Colorization CelebA DDRM FID 31.26 #1 of 3 Archive leaderboard report
Image Colorization CelebA DDNM Consistency 26.25 #2 of 3 Archive leaderboard report
Image Colorization CelebA DDNM FID 26.44 #2 of 3 Archive leaderboard report
Image Colorization CelebA A+y Consistency 0 #3 of 3 Archive leaderboard report
Image Colorization CelebA A+y FID 68.81 #3 of 3 Archive leaderboard report
Image Colorization ImageNet DDRM Consistency 260.4 #1 of 4 Archive leaderboard report
Image Colorization ImageNet DDRM FID 36.56 #1 of 4 Archive leaderboard report
Image Colorization ImageNet DDNM Consistency 42.32 #2 of 4 Archive leaderboard report
Image Colorization ImageNet DDNM FID 36.32 #2 of 4 Archive leaderboard report
Image Colorization ImageNet A+y Consistency 0 #3 of 4 Archive leaderboard report
Image Colorization ImageNet A+y FID 43.37 #3 of 4 Archive leaderboard report
Image Colorization ImageNet DGP FID 69.54 #4 of 4 Archive leaderboard report
Image Deblurring CelebA DDNM FID 1.41 #1 of 3 Archive leaderboard report
Image Deblurring CelebA DDNM PSNR 46.72 #1 of 3 Archive leaderboard report
Image Deblurring CelebA DDNM SSIM 0.996 #1 of 3 Archive leaderboard report
Image Deblurring CelebA DDRM FID 6.24 #2 of 3 Archive leaderboard report
Image Deblurring CelebA DDRM PSNR 43.07 #2 of 3 Archive leaderboard report
Image Deblurring CelebA DDRM SSIM 0.993 #2 of 3 Archive leaderboard report
Image Deblurring CelebA A+y FID 54.31 #3 of 3 Archive leaderboard report
Image Deblurring CelebA A+y PSNR 18.85 #3 of 3 Archive leaderboard report
Image Deblurring CelebA A+y SSIM 0.741 #3 of 3 Archive leaderboard report
Image Deblurring ImageNet DDNM FID 1.15 #1 of 3 Archive leaderboard report
Image Deblurring ImageNet DDNM PSNR 44.93 #1 of 3 Archive leaderboard report
Image Deblurring ImageNet DDNM SSIM 0.994 #1 of 3 Archive leaderboard report
Image Deblurring ImageNet DDRM FID 1.48 #2 of 3 Archive leaderboard report
Image Deblurring ImageNet DDRM PSNR 43.01 #2 of 3 Archive leaderboard report
Image Deblurring ImageNet DDRM SSIM 0.992 #2 of 3 Archive leaderboard report
Image Deblurring ImageNet A+y FID 55.42 #3 of 3 Archive leaderboard report
Image Deblurring ImageNet A+y PSNR 18.56 #3 of 3 Archive leaderboard report
Image Deblurring ImageNet A+y SSIM 0.6616 #3 of 3 Archive leaderboard report
Image Inpainting CelebA DDNM FID 4.54 #1 of 5 Archive leaderboard report
Image Inpainting CelebA DDNM PSNR 35.64 #1 of 5 Archive leaderboard report
Image Inpainting CelebA DDNM SSIM 0.982 #1 of 5 Archive leaderboard report
Image Inpainting CelebA DDRM FID 12.53 #2 of 5 Archive leaderboard report
Image Inpainting CelebA DDRM PSNR 34.79 #2 of 5 Archive leaderboard report
Image Inpainting CelebA DDRM SSIM 0.978 #2 of 5 Archive leaderboard report
Image Inpainting CelebA RePaint FID 14.19 #3 of 5 Archive leaderboard report
Image Inpainting CelebA RePaint PSNR 35.2 #3 of 5 Archive leaderboard report
Image Inpainting CelebA RePaint SSIM 0.981 #3 of 5 Archive leaderboard report
Image Inpainting CelebA A+y FID 181.56 #4 of 5 Archive leaderboard report
Image Inpainting CelebA A+y PSNR 15.57 #4 of 5 Archive leaderboard report
Image Inpainting CelebA A+y SSIM 0.809 #4 of 5 Archive leaderboard report
Image Inpainting ImageNet DDNM FID 3.89 #2 of 5 Archive leaderboard report
Image Inpainting ImageNet DDNM PSNR 32.06 #2 of 5 Archive leaderboard report
Image Inpainting ImageNet DDNM SSIM 0.968 #2 of 5 Archive leaderboard report
Image Inpainting ImageNet DDRM FID 4.82 #3 of 5 Archive leaderboard report
Image Inpainting ImageNet DDRM PSNR 31.73 #3 of 5 Archive leaderboard report
Image Inpainting ImageNet DDRM SSIM 0.966 #3 of 5 Archive leaderboard report
Image Inpainting ImageNet RePaint FID 12.31 #4 of 5 Archive leaderboard report
Image Inpainting ImageNet RePaint PSNR 31.87 #4 of 5 Archive leaderboard report
Image Inpainting ImageNet RePaint SSIM 0.963 #4 of 5 Archive leaderboard report
Image Inpainting ImageNet A+y FID 72.71 #5 of 5 Archive leaderboard report
Image Inpainting ImageNet A+y PSNR 14.52 #5 of 5 Archive leaderboard report
Image Inpainting ImageNet A+y SSIM 0.799 #5 of 5 Archive leaderboard report
Image Super-Resolution CelebA DDNM FID 22.27 #1 of 5 Archive leaderboard report
Image Super-Resolution CelebA DDNM PSNR 31.63 #1 of 5 Archive leaderboard report
Image Super-Resolution CelebA DDNM SSIM 0.945 #1 of 5 Archive leaderboard report
Image Super-Resolution CelebA ILVR FID 29.82 #2 of 5 Archive leaderboard report
Image Super-Resolution CelebA ILVR PSNR 31.59 #2 of 5 Archive leaderboard report
Image Super-Resolution CelebA ILVR SSIM 0.945 #2 of 5 Archive leaderboard report
Image Super-Resolution CelebA DDRM FID 31.04 #3 of 5 Archive leaderboard report
Image Super-Resolution CelebA DDRM PSNR 31.63 #3 of 5 Archive leaderboard report
Image Super-Resolution CelebA DDRM SSIM 0.945 #3 of 5 Archive leaderboard report
Image Super-Resolution CelebA PULSE FID 40.33 #4 of 5 Archive leaderboard report
Image Super-Resolution CelebA PULSE PSNR 22.74 #4 of 5 Archive leaderboard report
Image Super-Resolution CelebA PULSE SSIM 0.623 #4 of 5 Archive leaderboard report
Image Super-Resolution CelebA A+y FID 103.3 #5 of 5 Archive leaderboard report
Image Super-Resolution CelebA A+y PSNR 27.27 #5 of 5 Archive leaderboard report
Image Super-Resolution CelebA A+y SSIM 0.782 #5 of 5 Archive leaderboard report
Image Super-Resolution ImageNet DDNM FID 39.26 #2 of 6 Archive leaderboard report
Image Super-Resolution ImageNet DDNM PSNR 27.46 #2 of 6 Archive leaderboard report
Image Super-Resolution ImageNet DDNM SSIM 0.87 #2 of 6 Archive leaderboard report
Image Super-Resolution ImageNet DDRM FID 43.15 #3 of 6 Archive leaderboard report
Image Super-Resolution ImageNet DDRM PSNR 27.38 #3 of 6 Archive leaderboard report
Image Super-Resolution ImageNet DDRM SSIM 0.869 #3 of 6 Archive leaderboard report
Image Super-Resolution ImageNet ILVR FID 43.66 #4 of 6 Archive leaderboard report
Image Super-Resolution ImageNet ILVR PSNR 27.4 #4 of 6 Archive leaderboard report
Image Super-Resolution ImageNet ILVR SSIM 0.87 #4 of 6 Archive leaderboard report
Image Super-Resolution ImageNet DGP FID 64.34 #5 of 6 Archive leaderboard report
Image Super-Resolution ImageNet DGP PSNR 23.18 #5 of 6 Archive leaderboard report
Image Super-Resolution ImageNet DGP SSIM 0.798 #5 of 6 Archive leaderboard report
Image Super-Resolution ImageNet A+y FID 134.4 #6 of 6 Archive leaderboard report
Image Super-Resolution ImageNet A+y PSNR 24.26 #6 of 6 Archive leaderboard report
Image Super-Resolution ImageNet A+y SSIM 0.684 #6 of 6 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

Diffusion

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