Papers › Discovering Transferable Forensic Features for CNN-generated Images Detection

Discovering Transferable Forensic Features for CNN-generated Images Detection

24 Aug 2022arXiv:2208.11342archive 2025-07-28

Keshigeyan Chandrasegaran, Ngoc-Trung Tran, Alexander Binder, Ngai-Man Cheung

Visual counterfeits are increasingly causing an existential conundrum in mainstream media with rapid evolution in neural image synthesis methods. Though detection of such counterfeits has been a taxing problem in the image forensics community, a recent class of forensic detectors -- universal detectors -- are able to surprisingly spot counterfeit images regardless of generator architectures, loss functions, training datasets, and resolutions. This intriguing property suggests the possible existence of transferable forensic features (T-FF) in universal detectors. In this work, we conduct the first analytical study to discover and understand T-FF in universal detectors. Our contributions are 2-fold: 1) We propose a novel forensic feature relevance statistic (FF-RS) to quantify and discover T-FF in universal detectors and, 2) Our qualitative and quantitative investigations uncover an unexpected finding: color is a critical T-FF in universal detectors. Code and models are available at https://keshik6.github.io/transferable-forensic-features/

PaperPDFCodeCode 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 24 code samples harvested from 1 repository linked to this paper: 20 ran (of which 17 constructed an object rather than computing a result; 18 with no instrument failure: 0 honoured, 0 violated, 18 with no contract checked; 2 where Syntology's instrument failed) · 4 unverified. Where Syntology's instrument failed: 2 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 24 samples from 1 repository, 20 ran (18 with no instrument failure; 2 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.

sutd-visual-computing-group/transferable-forensic-features officialmentioned on GitHubpytorchSyntology: 24 samples, 20 ran (2 instrument-failed) · 4 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

24 samples harvested: 20 ran (of which 17 constructed an object rather than computing a result; 18 with no instrument failure: 0 honoured, 0 violated, 18 with no contract checked; 2 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.

2ran · our draft was wrong
18ran
4unverified

Licence: 0 of the 24 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 sutd-visual-computing-group/transferable-forensic-features. “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”. For 7 samples that ran, the graph records a signature this page does not show: for a class the fixture built without calling, the harness called the object with inputs it made up itself, so the recorded signature is the harness's probe, not the fixture's arguments. “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.

sutd-visual-computing-group/transferable-forensic-features official repository 24 samples, 20 ran (2 instrument-failed) · 4 unverified
anysign_conv src/lrp/ef_lrp_general.py ran MIT (permissive) · 1b10aaafa5ffab6a · report fixture: returned anysign_conv · level recorded 2026-09-22
clamplayer src/lrp/ef_lrp_general.py ran fingerprinted MIT (permissive) · 5c84599d73380ab1 · report fixture: returned clamplayer · level recorded 2026-09-22
conv2d_beta0_wrapper_fct src/lrp/ef_lrp_general.py ran MIT (permissive) · 821c88d2437ded8b · report fixture: returned conv2d_beta0_wrapper_fct · level recorded 2026-09-22
conv2d_betaadaptive_wrapper_fct src/lrp/ef_lrp_general.py ran MIT (permissive) · 0a9bf0583b9082f9 · report fixture: returned conv2d_betaadaptive_wrapper_fct · level recorded 2026-09-22
conv2d_betaany_wrapper_fct src/lrp/ef_lrp_general.py ran MIT (permissive) · 7f841ae57d02476e · report fixture: returned conv2d_betaany_wrapper_fct · level recorded 2026-09-22
conv2d_zbeta_wrapper_fct src/lrp/ef_lrp_general.py ran MIT (permissive) · 9e93c92453d76d41 · report fixture: returned conv2d_zbeta_wrapper_fct · level recorded 2026-09-22
invertedposnegconv src/lrp/ef_lrp_general.py ran fingerprinted MIT (permissive) · cb64f4447a9ba0f8 · report fixture: returned invertedposnegconv · level recorded 2026-09-22
lrp_backward src/lrp/ef_lrp_general.py ran · our draft was wrong MIT (permissive) · 50923b0f141f095a · report fixture: signature _input:Tensor/2/float, layer:callable/0/?, relevance_output:Tensor/2/float, eps0:float/0/float, eps:float/0/float · returned Tensor · level recorded 2026-09-22
lrplookupnotfounderror src/lrp/ef_lrp_general.py ran MIT (permissive) · 4f2c64e82470851e · report fixture: returned lrplookupnotfounderror · level recorded 2026-09-22
oneparam_wrapper_class src/lrp/ef_lrp_general.py ran MIT (permissive) · 4baeba022cff9de1 · report fixture: returned oneparam_wrapper_class · level recorded 2026-09-22
posnegconv src/lrp/ef_lrp_general.py ran fingerprinted MIT (permissive) · f927f8950ba9d466 · report fixture: returned posnegconv · level recorded 2026-09-22
safe_divide in 19 papers
census: 3 implementations from 5 papers, 2 distinct outputs on one shared input this sample: bucket 1 of 1, arg 1: rank 2, kind float, dtype float32 · arg 2: rank 2, kind float, dtype float32
src/lrp/ef_lrp_general.py ran · our draft was wrong MIT (permissive) · a215b24b1c0382ec · report fixture: signature numerator:Tensor/2/float, divisor:Tensor/2/float, eps0:float/0/float, eps:float/0/float · returned Tensor · level recorded 2026-09-22
sum_stacked2 src/lrp/ef_lrp_general.py ran fingerprinted MIT (permissive) · 8cbafe92b9230b91 · report fixture: signature x:Tensor/3/float · returned Tensor · level recorded 2026-09-24
tensorbiased_convlayer src/lrp/ef_lrp_general.py ran fingerprinted MIT (permissive) · db981f380a0a2297 · report fixture: returned tensorbiased_convlayer · level recorded 2026-09-22
tensorbiased_linearlayer src/lrp/ef_lrp_general.py ran fingerprinted MIT (permissive) · 3e42844a6a729935 · report fixture: returned tensorbiased_linearlayer · level recorded 2026-09-22
tensorbiasedconv2d_beta0_wrapper_fct src/lrp/ef_lrp_general.py ran MIT (permissive) · e38d338dc9d2c79e · report fixture: returned tensorbiasedconv2d_beta0_wrapper_fct · level recorded 2026-09-22
tensorbiasedconv2d_betaadaptive_wrapper_fct src/lrp/ef_lrp_general.py ran MIT (permissive) · a9191d22ad9fbd1d · report fixture: returned tensorbiasedconv2d_betaadaptive_wrapper_fct · level recorded 2026-09-22
tensorbiasedconv2d_betaany_wrapper_fct src/lrp/ef_lrp_general.py ran MIT (permissive) · cfb50acee5dfcbe7 · report fixture: returned tensorbiasedconv2d_betaany_wrapper_fct · level recorded 2026-09-22
twoparam_wrapper_class src/lrp/ef_lrp_general.py ran fingerprinted MIT (permissive) · 5b89691311760655 · report fixture: returned twoparam_wrapper_class · level recorded 2026-09-22
zeroparam_wrapper_class src/lrp/ef_lrp_general.py ran fingerprinted MIT (permissive) · 42cfb2a071642abe · report fixture: returned zeroparam_wrapper_class · level recorded 2026-09-22
conv2d_zbeta_wrapper_class src/lrp/ef_lrp_general.py unverified MIT (permissive) · bee51e3019e752b5 · report
get_lrpwrapperformodule src/lrp/ef_lrp_general.py unverified MIT (permissive) · 7b712b5d43f45b25 · report
invertedposnegconv_tensorbiased src/lrp/ef_lrp_general.py unverified MIT (permissive) · c96113c193c0bf5c · report
posnegconv_tensorbiased src/lrp/ef_lrp_general.py unverified MIT (permissive) · 57cc99f817a6e87b · report

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.

1 assembly on this paper: 1 harvested, 0 generated.

harvested Assembly for FF-RS (graph method: FF-RS). 24 bricks; individually on a fixture: 20 ran (of which 17 constructed an object rather than computing a result; 18 with no instrument failure: 0 honoured, 0 violated, 18 with no contract checked; 2 where Syntology's instrument failed) · 4 unverified. The assembly has not been run end-to-end.

  1. get_lrpwrapperformodule entry · level 0 · unverified · 7b712b5d43f45b25
  2. clamplayer helper · level 1 · ran · 5c84599d73380ab1
  3. conv2d_beta0_wrapper_fct model · level 1 · ran · 821c88d2437ded8b
  4. conv2d_betaadaptive_wrapper_fct model · level 1 · ran · 0a9bf0583b9082f9
  5. conv2d_betaany_wrapper_fct model · level 1 · ran · 7f841ae57d02476e
  6. conv2d_zbeta_wrapper_class model · level 0 · unverified · bee51e3019e752b5
  7. lrplookupnotfounderror helper · level 1 · ran · 4f2c64e82470851e
  8. oneparam_wrapper_class helper · level 1 · ran · 4baeba022cff9de1
  9. sum_stacked2 helper · level 1 · ran · 8cbafe92b9230b91
  10. tensorbiased_convlayer helper · level 1 · ran · db981f380a0a2297
  11. tensorbiased_linearlayer helper · level 1 · ran · 3e42844a6a729935
  12. tensorbiasedconv2d_beta0_wrapper_fct model · level 1 · ran · e38d338dc9d2c79e
  13. tensorbiasedconv2d_betaadaptive_wrapper_fct model · level 1 · ran · a9191d22ad9fbd1d
  14. tensorbiasedconv2d_betaany_wrapper_fct model · level 1 · ran · cfb50acee5dfcbe7
  15. twoparam_wrapper_class helper · level 1 · ran · 5b89691311760655
  16. zeroparam_wrapper_class helper · level 1 · ran · 42cfb2a071642abe
  17. lrp_backward model · level 1 · ran · our draft was wrong · 50923b0f141f095a
  18. posnegconv model · level 1 · ran · f927f8950ba9d466
  19. invertedposnegconv model · level 1 · ran · cb64f4447a9ba0f8
  20. conv2d_zbeta_wrapper_fct model · level 1 · ran · 9e93c92453d76d41
  21. posnegconv_tensorbiased model · level 0 · unverified · 57cc99f817a6e87b
  22. invertedposnegconv_tensorbiased model · level 0 · unverified · c96113c193c0bf5c
  23. safe_divide helper · level 1 · ran · our draft was wrong · a215b24b1c0382ec
  24. anysign_conv model · level 1 · ran · 1b10aaafa5ffab6a

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="2208.11342") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Image ForensicsImage Generation

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

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