Papers › Counterfactual Attention Learning for Fine-Grained Visual Categorization and Re-identification

Counterfactual Attention Learning for Fine-Grained Visual Categorization and Re-identification

19 Aug 2021ICCV 2021 10arXiv:2108.08728archive 2025-07-28

Yongming Rao, Guangyi Chen, Jiwen Lu, Jie zhou

Attention mechanism has demonstrated great potential in fine-grained visual recognition tasks. In this paper, we present a counterfactual attention learning method to learn more effective attention based on causal inference. Unlike most existing methods that learn visual attention based on conventional likelihood, we propose to learn the attention with counterfactual causality, which provides a tool to measure the attention quality and a powerful supervisory signal to guide the learning process. Specifically, we analyze the effect of the learned visual attention on network prediction through counterfactual intervention and maximize the effect to encourage the network to learn more useful attention for fine-grained image recognition. Empirically, we evaluate our method on a wide range of fine-grained recognition tasks where attention plays a crucial role, including fine-grained image categorization, person re-identification, and vehicle re-identification. The consistent improvement on all benchmarks demonstrates the effectiveness of our method. Code is available at https://github.com/raoyongming/CAL

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 11 code samples harvested from 1 repository linked to this paper: 8 ran (of which 7 constructed an object rather than computing a result; 8 with no instrument failure: 0 honoured, 0 violated, 8 with no contract checked; 0 where Syntology's instrument failed) · 3 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code; here there are none.

By repository: official repository: 11 samples from 1 repository, 8 ran (8 with no instrument failure; 0 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.

raoyongming/CAL officialmentioned in papermentioned on GitHubpytorchSyntology: 11 samples, 8 ran (0 instrument-failed) · 3 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

11 samples harvested: 8 ran (of which 7 constructed an object rather than computing a result; 8 with no instrument failure: 0 honoured, 0 violated, 8 with no contract checked; 0 where Syntology's instrument failed) · 3 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.

8ran
3unverified

Licence: 0 of the 11 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 raoyongming/CAL. “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 6 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.

raoyongming/CAL official repository 11 samples, 8 ran (0 instrument-failed) · 3 unverified
BAP fgvc/models/cal.py ran fingerprinted MIT (permissive) · 003486e56cfd2870 · report fixture: returned BAP · level recorded 2026-09-22
BasicConv2d in 12 papers fgvc/models/cal.py ran fingerprinted MIT (permissive) · ee5a84cd4d4e6a4f · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-24
InceptionA fgvc/models/cal.py ran fingerprinted MIT (permissive) · 32e7d9b157d87237 · report fixture: returned InceptionA · level recorded 2026-09-22
InceptionAux fgvc/models/cal.py ran MIT (permissive) · 9e5dae472f33d1cb · report fixture: returned InceptionAux · level recorded 2026-09-22
InceptionB fgvc/models/cal.py ran fingerprinted MIT (permissive) · 1ed069706334c6a4 · report fixture: returned InceptionB · level recorded 2026-09-22
InceptionC fgvc/models/cal.py ran fingerprinted MIT (permissive) · be3a20ab4edbbfff · report fixture: returned InceptionC · level recorded 2026-09-22
InceptionD fgvc/models/cal.py ran fingerprinted MIT (permissive) · ca178667076da355 · report fixture: returned InceptionD · level recorded 2026-09-22
InceptionE fgvc/models/cal.py ran fingerprinted MIT (permissive) · 6d1c24d8babf109b · report fixture: returned InceptionE · level recorded 2026-09-22
Inception3 fgvc/models/cal.py unverified MIT (permissive) · bf7a0b9b99f8c230 · report
WSDAN_CAL fgvc/models/cal.py unverified MIT (permissive) · 00e4a6e5da3d2d97 · report
inception_v3 in 16 papers fgvc/models/cal.py unverified MIT (permissive) · 5113305f69f8a813 · 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 Counterfactual Attention Learning (graph method: Counterfactual Attention Learning). 11 bricks; individually on a fixture: 8 ran (of which 7 constructed an object rather than computing a result; 8 with no instrument failure: 0 honoured, 0 violated, 8 with no contract checked; 0 where Syntology's instrument failed) · 3 unverified. The assembly has not been run end-to-end.

  1. WSDAN_CAL entry · level 0 · unverified · 00e4a6e5da3d2d97
  2. BAP model · level 1 · ran · 003486e56cfd2870
  3. BasicConv2d helper · level 1 · ran · ee5a84cd4d4e6a4f
  4. inception_v3 helper · level 0 · unverified · 5113305f69f8a813
  5. Inception3 helper · level 0 · unverified · bf7a0b9b99f8c230
  6. InceptionA helper · level 1 · ran · 32e7d9b157d87237
  7. InceptionAux helper · level 1 · ran · 9e5dae472f33d1cb
  8. InceptionB helper · level 1 · ran · 1ed069706334c6a4
  9. InceptionC helper · level 1 · ran · be3a20ab4edbbfff
  10. InceptionD helper · level 1 · ran · ca178667076da355
  11. InceptionE helper · level 1 · ran · 6d1c24d8babf109b

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="2108.08728") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Causal InferenceFew-Shot LearningFine-Grained Image ClassificationFine-Grained Image RecognitionFine-Grained Visual CategorizationFine-Grained Visual RecognitionImage CategorizationMitigating Contextual BiasPerson Re-IdentificationVehicle Re-Identification

1 archive task tag without a task page not shown.

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Few-Shot Learning DTD CAL 12-shot Accuracy 54.6 #3 of 4 Archive leaderboard report
Few-Shot Learning DTD CAL 16-shot Accuracy 57.4 #3 of 4 Archive leaderboard report
Few-Shot Learning DTD CAL 4-shot Accuracy 40.9 #3 of 4 Archive leaderboard report
Few-Shot Learning DTD CAL 8-shot Accuracy 50.4 #3 of 4 Archive leaderboard report
Few-Shot Learning FGVC Aircraft CAL 12-shot Accuracy 67.6 #2 of 4 Archive leaderboard report
Few-Shot Learning FGVC Aircraft CAL 16-shot Accuracy 74.3 #2 of 4 Archive leaderboard report
Few-Shot Learning FGVC Aircraft CAL 4-shot Accuracy 35.2 #2 of 4 Archive leaderboard report
Few-Shot Learning FGVC Aircraft CAL 8-shot Accuracy 55.4 #2 of 4 Archive leaderboard report
Few-Shot Learning FGVC Aircraft CAL Harmonic mean 35.2 #2 of 4 Archive leaderboard report
Few-Shot Learning Stanford Cars CAL 12-shot Accuracy 82.9 #3 of 3 Archive leaderboard report
Few-Shot Learning Stanford Cars CAL 16-shot Accuracy 88.9 #3 of 3 Archive leaderboard report
Few-Shot Learning Stanford Cars CAL 4-shot Accuracy 42.2 #3 of 3 Archive leaderboard report
Few-Shot Learning Stanford Cars CAL 8-shot Accuracy 71.8 #3 of 3 Archive leaderboard report
Fine-Grained Image Classification CUB-200-2011 CAL Accuracy 90.6 #12 of 30 Archive leaderboard report
Fine-Grained Image Classification FGVC Aircraft CAL Accuracy 94.2 #11 of 57 Archive leaderboard report
Fine-Grained Image Classification Stanford Cars CAL Accuracy 95.5% #13 of 83 Archive leaderboard report
Mitigating Contextual Bias FGVC Aircraft CAL + ALIA OOD Accuracy (%) 25.1 #2 of 4 Archive leaderboard report
Mitigating Contextual Bias FGVC Aircraft CAL + ALIA Top-1 Accuracy (%) 71.8 #2 of 4 Archive leaderboard report
Mitigating Contextual Bias FGVC Aircraft CAL OOD Accuracy (%) 10.2 #4 of 4 Archive leaderboard report
Mitigating Contextual Bias FGVC Aircraft CAL Top-1 Accuracy (%) 71.0 #4 of 4 Archive leaderboard report
Person Re-Identification DukeMTMC-reID CAL Rank-1 90 #41 of 94 Archive leaderboard report
Person Re-Identification DukeMTMC-reID CAL mAP 80.5 #41 of 94 Archive leaderboard report
Person Re-Identification MSMT17 CAL(ResNet50) Rank-1 84.2 #24 of 43 Archive leaderboard report
Person Re-Identification MSMT17 CAL(ResNet50) mAP 64 #24 of 43 Archive leaderboard report
Person Re-Identification Market-1501 CAL Rank-1 95.5 #55 of 135 Archive leaderboard report
Person Re-Identification Market-1501 CAL mAP 89.5 #55 of 135 Archive leaderboard report
Vehicle Re-Identification VeRi-776 CAL Rank-1 95.4 #15 of 17 Archive leaderboard report
Vehicle Re-Identification VeRi-776 CAL Rank5 97.9 #15 of 17 Archive leaderboard report
Vehicle Re-Identification VeRi-776 CAL mAP 74.3 #15 of 17 Archive leaderboard report
Vehicle Re-Identification VehicleID Large CAL Rank-1 75.1 #9 of 10 Archive leaderboard report
Vehicle Re-Identification VehicleID Large CAL mAP 80.9 #9 of 10 Archive leaderboard report
Vehicle Re-Identification VehicleID Medium CAL Rank-1 78.2 #8 of 9 Archive leaderboard report
Vehicle Re-Identification VehicleID Medium CAL mAP 83.8 #8 of 9 Archive leaderboard report
Vehicle Re-Identification VehicleID Small CAL Rank-1 82.5 #10 of 13 Archive leaderboard report
Vehicle Re-Identification VehicleID Small CAL mAP 87.8 #10 of 13 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