Papers › Learning to Generate Realistic Noisy Images via Pixel-level Noise-aware Adversarial Training

Learning to Generate Realistic Noisy Images via Pixel-level Noise-aware Adversarial Training

6 Apr 2022NeurIPS 2021 12arXiv:2204.02844archive 2025-07-28

Yuanhao Cai, Xiaowan Hu, Haoqian Wang, Yulun Zhang, Hanspeter Pfister, Donglai Wei

Existing deep learning real denoising methods require a large amount of noisy-clean image pairs for supervision. Nonetheless, capturing a real noisy-clean dataset is an unacceptable expensive and cumbersome procedure. To alleviate this problem, this work investigates how to generate realistic noisy images. Firstly, we formulate a simple yet reasonable noise model that treats each real noisy pixel as a random variable. This model splits the noisy image generation problem into two sub-problems: image domain alignment and noise domain alignment. Subsequently, we propose a novel framework, namely Pixel-level Noise-aware Generative Adversarial Network (PNGAN). PNGAN employs a pre-trained real denoiser to map the fake and real noisy images into a nearly noise-free solution space to perform image domain alignment. Simultaneously, PNGAN establishes a pixel-level adversarial training to conduct noise domain alignment. Additionally, for better noise fitting, we present an efficient architecture Simple Multi-scale Network (SMNet) as the generator. Qualitative validation shows that noise generated by PNGAN is highly similar to real noise in terms of intensity and distribution. Quantitative experiments demonstrate that a series of denoisers trained with the generated noisy images achieve state-of-the-art (SOTA) results on four real denoising benchmarks. Part of codes, pre-trained models, and results are available at https://github.com/caiyuanhao1998/PNGAN for comparisons.

PaperPDFConference PDFCodeCode Syntology ranMethod as code

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

Syntology 18 code samples harvested from 2 repositories linked to this paper: 14 ran (of which 0 constructed an object rather than computing a result; 13 with no instrument failure: 0 honoured, 0 violated, 13 with no contract checked; 1 where Syntology's instrument failed) · 4 unverified. Where Syntology's instrument failed: 1 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: named in the paper: 12 samples from 1 repository, 8 ran (8 with no instrument failure; 0 where Syntology's instrument failed); community (archive-listed): 6 samples from 1 repository, 6 ran (5 with no instrument failure; 1 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.

Levels written between 2026-09-22 and 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

caiyuanhao1998/PNGAN mentioned in paperpytorchSyntology: 12 samples, 8 ran (0 instrument-failed) · 4 unverified report
GarrickZ2/Image-Denoising pytorchSyntology: 6 samples, 6 ran (1 instrument-failed) · 0 unverified 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

18 samples harvested: 14 ran (of which 0 constructed an object rather than computing a result; 13 with no instrument failure: 0 honoured, 0 violated, 13 with no contract checked; 1 where Syntology's instrument failed) · 4 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

1ran · our draft was wrong
13ran
4unverified

Licence: 0 of the 18 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 2 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. “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.

caiyuanhao1998/PNGAN named in the paper 12 samples, 8 ran (0 instrument-failed) · 4 unverified
DAU networks/MIRNet_model.py ran MIT (permissive) · 98fb983c7e88720b · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-24
DownSample in 11 papers networks/MIRNet_model.py ran fingerprinted MIT (permissive) · 38aa16cfac60cc92 · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-24
ResidualDownSample networks/MIRNet_model.py ran fingerprinted MIT (permissive) · cdc02a8705d4e515 · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-24
SKFF networks/MIRNet_model.py ran · metamorphic tier: well formed MIT (permissive) · edb5af10fa50176e · report fixture: signature inp_feats:list/1/Tensor · returned Tensor · level recorded 2026-09-22
UpSample in 9 papers networks/MIRNet_model.py ran · metamorphic tier: invariant fingerprinted MIT (permissive) · 84169db3c61ab293 · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-22
ca_layer networks/MIRNet_model.py ran · metamorphic tier: invariant MIT (permissive) · 6379151f00888f20 · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-22
downsamp networks/MIRNet_model.py ran · metamorphic tier: invariant fingerprinted MIT (permissive) · bf2eb3936cb10010 · report fixture: signature inp:Tensor/4/float · returned Tensor · level recorded 2026-09-22
spatial_attn_layer networks/MIRNet_model.py ran · metamorphic tier: invariant fingerprinted MIT (permissive) · 0432f3b4007a10d2 · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-22
MIRNet networks/MIRNet_model.py unverified MIT (permissive) · 82219d825869b542 · report
MSRB networks/MIRNet_model.py unverified MIT (permissive) · 67a302b3bbe19d09 · report
RRG networks/MIRNet_model.py unverified MIT (permissive) · 9ddd23ff4f39e64c · report
ResidualUpSample networks/MIRNet_model.py unverified MIT (permissive) · b6ff14afe822f838 · report
GarrickZ2/Image-Denoising community (archive-listed) 6 samples, 6 ran (1 instrument-failed) · 0 unverified
BlurPool2d PNGAN/model/generator.py ran · metamorphic tier: invariant fingerprinted MIT (permissive) · 9cf54f0737819c78 · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-22
FCA PNGAN/model/generator.py ran · metamorphic tier: invariant fingerprinted MIT (permissive) · f1293d5b242d5302 · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-22
Generator in 95 papers PNGAN/model/generator.py ran fingerprinted MIT (permissive) · 88e02cc50ac1629c · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-24
MAB in 12 papers PNGAN/model/generator.py ran MIT (permissive) · 0328887c18f80d1f · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-24
SRG PNGAN/model/generator.py ran MIT (permissive) · 94c45fe09610c1f3 · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-24
up_sample PNGAN/model/generator.py ran · our draft was wrong MIT (permissive) · d133dd217f98b88c · report fixture: signature channels:int/0/int, scale:int/0/int, k_size:int/0/int, padding:int/0/int · returned Sequential · level recorded 2026-09-22

Method as code Syntology

What Syntology's graph records as this paper's claimed method, lifted into code: each assembly names its bricks (functions and classes), each with its own role and its own verification level. An assembly is not a check that the code implements the method correctly or reproduces the paper's results. Levels are each brick's own verification level in Syntology's graph (level 0: no recorded run), listed brick by brick and never combined into one level; “ran” means the brick executed on a synthesized input, not that its output is correct, and the instrument figure counts failures of Syntology's instrument, not of the code. Where a model read the code against the paper's claim, each assembly's label verdict is one model's judgement, kept in this page's JSON twin; it is not calibrated, so no card prints it. Siblings are assemblies lifted from this paper in the same batch for the same method: alternative readings kept side by side; none replaces or supersedes another.

2 assemblies on this paper: 2 harvested, 0 generated, listed in 1 groups (siblings lifted in one batch form one group).

2 sibling assemblies for Pixel-level Noise-aware Generative Adversarial Network (PNGAN) (harvested), lifted in the same batch on 2026-09-09: alternative readings kept side by side; none replaces or supersedes another.

harvested Assembly for Pixel-level Noise-aware Generative Adversarial Network (PNGAN) (graph method: Pixel-level Noise-aware Generative Adversarial Network (PNGAN)). 7 bricks; individually on a fixture: 7 ran (of which 0 constructed an object rather than computing a result; 6 with no instrument failure: 1 honoured, 0 violated, 5 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified. The assembly has not been run end-to-end.

one of 2 siblings

  1. Generator entry · level 1 · ran · 88e02cc50ac1629c
  2. SRG model · level 1 · ran · 94c45fe09610c1f3
  3. MAB model · level 1 · ran · 0328887c18f80d1f
  4. BlurPool2d helper · level 1 · ran · 9cf54f0737819c78
  5. FCA helper · level 1 · ran · f1293d5b242d5302
  6. up_sample helper · level 1 · ran · our draft was wrong · d133dd217f98b88c
  7. get_padding helper · level 1 · ran · honoured contract · 73876e077d3acdf5

harvested Assembly for Pixel-level Noise-aware Generative Adversarial Network (PNGAN) (graph method: Pixel-level Noise-aware Generative Adversarial Network (PNGAN)). 16 bricks; individually on a fixture: 12 ran (of which 1 constructed an object rather than computing a result; 10 with no instrument failure: 0 honoured, 0 violated, 10 with no contract checked; 2 where Syntology's instrument failed) · 4 unverified. The assembly has not been run end-to-end.

one of 2 siblings

  1. MIRNet entry · level 0 · unverified · 82219d825869b542
  2. RRG model · level 0 · unverified · 9ddd23ff4f39e64c
  3. MSRB model · level 0 · unverified · 67a302b3bbe19d09
  4. conv helper · level 1 · ran · our draft was wrong · 013ca1b768620edd
  5. DAU model · level 1 · ran · 98fb983c7e88720b
  6. DownSample helper · level 1 · ran · 38aa16cfac60cc92
  7. SKFF helper · level 1 · ran · edb5af10fa50176e
  8. UpSample helper · level 1 · ran · 84169db3c61ab293
  9. ca_layer model · level 1 · ran · 6379151f00888f20
  10. spatial_attn_layer model · level 1 · ran · 0432f3b4007a10d2
  11. ResidualDownSample helper · level 1 · ran · cdc02a8705d4e515
  12. ResidualUpSample helper · level 0 · unverified · b6ff14afe822f838
  13. BasicConv helper · level 1 · ran · 704f1adafa5b1dc0
  14. ChannelPool helper · level 1 · ran · 436816a2dc1cce55
  15. downsamp helper · level 1 · ran · bf2eb3936cb10010
  16. get_pad_layer helper · level 1 · ran · our draft was wrong · 50aa4a59fe718c61

Each brick ends with its code_sha256, Syntology's identity for that exact code; click one to copy the get_code call that returns it. For agents, on Syntology's MCP service: assembly(arxiv_id="2204.02844") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

DenoisingImage DenoisingImage GenerationNoise Estimation

1 archive task tag without a task page not shown.

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Image Denoising DND PNGAN PSNR (sRGB) 40.18 #2 of 16 Archive leaderboard report
Image Denoising DND PNGAN SSIM (sRGB) 0.961 #2 of 16 Archive leaderboard report
Image Denoising Nam PNGAN PSNR 40.78 #1 of 1 Archive leaderboard report
Image Denoising Nam PNGAN SSIM 0.986 #1 of 1 Archive leaderboard report
Image Denoising PolyU PNGAN PSNR 40.55 #1 of 1 Archive leaderboard report
Image Denoising PolyU PNGAN SSIM 0.983 #1 of 1 Archive leaderboard report
Image Denoising SIDD PNGAN PSNR (sRGB) 40.07 #5 of 22 Archive leaderboard report
Image Denoising SIDD PNGAN SSIM (sRGB) 0.960 #5 of 22 Archive leaderboard report
Noise Estimation SIDD PNGAN Average KL Divergence 0.153 #1 of 5 Archive leaderboard report
Noise Estimation SIDD PNGAN PSNR Gap 0.84 #1 of 5 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