Papers › AdaFace: Quality Adaptive Margin for Face Recognition

AdaFace: Quality Adaptive Margin for Face Recognition

3 Apr 2022CVPR 2022 1arXiv:2204.00964archive 2025-07-28

Minchul Kim, Anil K. Jain, Xiaoming Liu

Recognition in low quality face datasets is challenging because facial attributes are obscured and degraded. Advances in margin-based loss functions have resulted in enhanced discriminability of faces in the embedding space. Further, previous studies have studied the effect of adaptive losses to assign more importance to misclassified (hard) examples. In this work, we introduce another aspect of adaptiveness in the loss function, namely the image quality. We argue that the strategy to emphasize misclassified samples should be adjusted according to their image quality. Specifically, the relative importance of easy or hard samples should be based on the sample's image quality. We propose a new loss function that emphasizes samples of different difficulties based on their image quality. Our method achieves this in the form of an adaptive margin function by approximating the image quality with feature norms. Extensive experiments show that our method, AdaFace, improves the face recognition performance over the state-of-the-art (SoTA) on four datasets (IJB-B, IJB-C, IJB-S and TinyFace). Code and models are released in https://github.com/mk-minchul/AdaFace.

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 22 code samples harvested from 6 repositories linked to this paper: 17 ran (of which 2 constructed an object rather than computing a result; 14 with no instrument failure: 2 honoured, 1 violated, 11 with no contract checked; 3 where Syntology's instrument failed) · 5 unverified. Where Syntology's instrument failed: 3 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 2 samples from 1 repository, 2 ran (2 with no instrument failure; 0 where Syntology's instrument failed); community (archive-listed): 20 samples from 5 repositories, 15 ran (12 with no instrument failure; 3 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 1 of its repositories that Syntology harvested for other papers:

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.

mk-minchul/adaface officialmentioned in papermentioned on GitHubpytorchMITSyntology: 2 samples, 2 ran (0 instrument-failed) · 0 unverified report
chelsea234/m2f2_det mentioned on GitHubpytorchMITSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Rethinking Vision-Language Model in Face Forensics: Multi-Modal Interpretable Forged... (arXiv:2503.20188; that paper's own run record: 9 ran (of which 0 constructed an object rather than computing a result; 7 with no instrument failure: 1 honoured, 1 violated, 5 with no contract checked; 2 where Syntology's instrument failed) · 2 unverified); nothing checks that they implement this paper's method. report
sithu31296/EasyFace mentioned on GitHubpytorchSyntology: 2 samples, 2 ran (1 instrument-failed) · 0 unverified report
tomas-gajarsky/facetorch mentioned on GitHubpytorchSyntology: 2 samples, 2 ran (0 instrument-failed) · 0 unverified report
Armxyz1/Results-on-RFW pytorchSyntology: 10 samples, 7 ran (0 instrument-failed) · 3 unverified report
RocketFlash/easy_metric_learning pytorchSyntology: 4 samples, 4 ran (2 instrument-failed) · 0 unverified report
leondgarse/Keras_insightface tfSyntology: 2 samples, 0 ran · 2 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.

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

22 samples harvested: 17 ran (of which 2 constructed an object rather than computing a result; 14 with no instrument failure: 2 honoured, 1 violated, 11 with no contract checked; 3 where Syntology's instrument failed) · 5 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 · honoured contract
1ran · violated contract
3ran · our draft was wrong
11ran
5unverified

Licence: 11 of the 22 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 6 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. 1 sample whose link to this paper records no repository is listed under one of this paper's own repositories, because identical code was harvested from it for another 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”. For 1 sample 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.

mk-minchul/adaface official repository 2 samples, 2 ran (0 instrument-failed) · 0 unverified
AdaFace head.py ran MIT (permissive) · 775e2b18a7335c45 · report fixture: signature embbedings:Tensor/2/float, norms:Tensor/2/float, label:Tensor/1/int · returned Tensor · level recorded 2026-09-24
get_block in 40 papers net.py identical code Syntology harvested from this repository for another paper; this paper's link records no repository, path or licence of its own ran · honoured contract fingerprinted licence of this copy not recorded · ad5dbf57f3ea2633 · report fixture: signature in_channel:int/0/int, depth:int/0/int, num_units:int/0/int, stride:int/0/int · returned list · level recorded 2026-09-22
sithu31296/EasyFace community (archive-listed) 2 samples, 2 ran (1 instrument-failed) · 0 unverified
AdaFace easyface/recognition/heads/adaface.py ran MIT (permissive) · fdd08128d4e32cda · report fixture: signature feats:Tensor/2/float, norms:Tensor/2/float, label:Tensor/1/int · returned Tensor · level recorded 2026-09-24
l2_norm in 71 papers
census: 17 implementations from 25 papers, 8 distinct outputs across 3 buckets, one shared input per bucket this sample: bucket 1 of 3, arg 1: rank 2, kind float, dtype float32
easyface/recognition/heads/adaface.py ran · our draft was wrong fingerprinted MIT (permissive) · 32e701066e89cd01 · report fixture: signature x:Tensor/2/float, dim:int/0/int · returned Tensor · level recorded 2026-09-22
tomas-gajarsky/facetorch community (archive-listed) 2 samples, 2 ran (0 instrument-failed) · 0 unverified
MagFaceIResNet100 model_defs/verify_model.py ran · metamorphic tier: deterministic Apache-2.0 (permissive) · 28bfa34e142cf7bf · report fixture: returned MagFaceIResNet100 · level recorded 2026-09-22
_MagFaceIBasicBlock model_defs/verify_model.py ran · metamorphic tier: deterministic fingerprinted Apache-2.0 (permissive) · 1d1184e941b3a490 · report fixture: returned _MagFaceIBasicBlock · level recorded 2026-09-22
Armxyz1/Results-on-RFW community (archive-listed) 10 samples, 7 ran (0 instrument-failed) · 3 unverified
BasicBlockIR AdaFace/backbones/net.py ran · metamorphic tier: invariant fingerprinted no licence file found · pointer only · d9cbf9097e9b2f3b · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-22
BasicBlockIRSE AdaFace/backbones/net.py ran no licence file found · pointer only · c7ff115d76b10e48 · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-24
BottleneckIR AdaFace/backbones/net.py ran · metamorphic tier: invariant fingerprinted no licence file found · pointer only · 5ff2d0550242806c · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-22
BottleneckIRSE AdaFace/backbones/net.py ran · metamorphic tier: invariant no licence file found · pointer only · e880b0d152ae66fd · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-22
Flatten in 23 papers
census: 9 implementations from 9 papers, 2 distinct outputs across 2 buckets, one shared input per bucket this sample: bucket 2 of 2, arg 1: rank 3, kind float, dtype float32
AdaFace/backbones/net.py ran · metamorphic tier: invariant fingerprinted no licence file found · pointer only · 3f852d5b1c843b52 · report fixture: signature input:Tensor/3/float · returned Tensor · level recorded 2026-09-22
SEModule in 7 papers AdaFace/backbones/net.py ran · metamorphic tier: invariant fingerprinted no licence file found · pointer only · 20b077f2201866aa · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-22
get_blocks in 39 papers AdaFace/backbones/net.py ran · honoured contract no licence file found · pointer only · e82dfb7d2386721b · report fixture: signature num_layers:int/0/int · returned list · level recorded 2026-09-22
Backbone in 17 papers AdaFace/backbones/net.py unverified no licence file found · pointer only · f3ef8c1d9f566160 · report
Bottleneck in 98 papers AdaFace/backbones/net.py unverified no licence file found · pointer only · c9258e60d2cf27ee · report
initialize_weights in 30 papers AdaFace/backbones/net.py unverified no licence file found · pointer only · dd1764dcd0628ab8 · report
RocketFlash/easy_metric_learning community (archive-listed) 4 samples, 4 ran (2 instrument-failed) · 0 unverified
AdaFace src/model/margin/adaface.py ran Apache-2.0 (permissive) · 25446533fe6fcf5d · report fixture: signature x:Tensor/2/float, label:Tensor/1/int · returned Tensor · level recorded 2026-09-24
build_one_hot src/model/margin/adaface.py ran · our draft was wrong Apache-2.0 (permissive) · 00fcae2b28b19dbb · report fixture: signature label:Tensor/1/int, num_classes:int/0/int, device:str/0/str, dtype:dtype/0/? · returned Tensor · level recorded 2026-09-22
is_mixed_label src/model/margin/adaface.py ran · violated contract fingerprinted Apache-2.0 (permissive) · ef4ec1a6ba91a991 · report fixture: signature label:list/1/int · returned bool · level recorded 2026-09-22
l2_norm in 71 papers
census: 17 implementations from 25 papers, 8 distinct outputs across 3 buckets, one shared input per bucket this sample: bucket 1 of 3, arg 1: rank 2, kind float, dtype float32
src/model/margin/adaface.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 27524a6bac90b755 · report fixture: signature input_x:Tensor/2/float, axis:int/0/int · returned Tensor · level recorded 2026-09-22
leondgarse/Keras_insightface community (archive-listed) 2 samples, 0 ran · 2 unverified
AdaFaceLoss losses.py unverified MIT (permissive) · bf12a79527a79d4a · report
ArcfaceLossSimple losses.py unverified MIT (permissive) · 40d45aa734161334 · 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. 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.

6 assemblies on this paper: 6 harvested, 0 generated, listed in 1 groups (siblings lifted in one batch form one group).

6 sibling assemblies for AdaFace (harvested), lifted in the same batch on 2026-09-09: alternative readings kept side by side; none replaces or supersedes another.

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

one of 6 siblings

  1. MagFaceIResNet100 entry · level 1 · ran · 28bfa34e142cf7bf
  2. _MagFaceIBasicBlock model · level 1 · ran · 1d1184e941b3a490

harvested Assembly for AdaFace (graph method: AdaFace). 2 bricks; individually on a fixture: 0 ran · 2 unverified. The assembly has not been run end-to-end.

one of 6 siblings

  1. AdaFaceLoss entry · level 0 · unverified · bf12a79527a79d4a
  2. ArcfaceLossSimple helper · level 0 · unverified · 40d45aa734161334

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

one of 6 siblings

  1. AdaFace entry · level 1 · ran · fdd08128d4e32cda
  2. l2_norm helper · level 1 · ran · our draft was wrong · 32e701066e89cd01

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

one of 6 siblings

  1. AdaFace entry · level 1 · ran · 25446533fe6fcf5d
  2. build_one_hot helper · level 1 · ran · our draft was wrong · 00fcae2b28b19dbb
  3. l2_norm helper · level 1 · ran · our draft was wrong · 27524a6bac90b755
  4. is_mixed_label helper · level 1 · ran · violated contract · ef4ec1a6ba91a991

harvested Assembly for AdaFace (graph method: AdaFace). 11 bricks; individually on a fixture: 8 ran (of which 0 constructed an object rather than computing a result; 8 with no instrument failure: 2 honoured, 0 violated, 6 with no contract checked; 0 where Syntology's instrument failed) · 3 unverified. The assembly has not been run end-to-end.

one of 6 siblings

  1. Backbone entry · level 0 · unverified · f3ef8c1d9f566160
  2. BasicBlockIR model · level 1 · ran · d9cbf9097e9b2f3b
  3. BasicBlockIRSE helper · level 1 · ran · c7ff115d76b10e48
  4. BottleneckIR model · level 1 · ran · 5ff2d0550242806c
  5. BottleneckIRSE helper · level 1 · ran · e880b0d152ae66fd
  6. Flatten helper · level 1 · ran · 3f852d5b1c843b52
  7. get_blocks helper · level 1 · ran · honoured contract · e82dfb7d2386721b
  8. initialize_weights helper · level 0 · unverified · dd1764dcd0628ab8
  9. SEModule model · level 1 · ran · 20b077f2201866aa
  10. get_block helper · level 1 · ran · honoured contract · ad5dbf57f3ea2633
  11. Bottleneck helper · level 0 · unverified · c9258e60d2cf27ee

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

one of 6 siblings

  1. AdaFace entry · level 1 · ran · 775e2b18a7335c45
  2. l2_norm helper · level 1 · ran · our draft was wrong · c54fea429589425d

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="2204.00964") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Face RecognitionFace Recognition (Closed-Set)Face VerificationSurveillance-to-BookingSurveillance-to-SingleSurveillance-to-Surveillance

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Face Recognition IJB-B ArcFace + MS1MV2 + R100 Rank-1 0.9450 #2 of 4 Archive leaderboard report
Face Recognition IJB-B ArcFace + MS1MV2 + R100 TAR @ FAR=1e-5 0.8933 #2 of 4 Archive leaderboard report
Face Recognition IJB-B AdaFace + MS1MV3 + R100 TAR @ FAR=0.0001 0.9425 #4 of 4 Archive leaderboard report
Face Recognition LFW ArcFace + MS1MV2 + R100 Accuracy 0.9983 #5 of 16 Archive leaderboard report
Face Recognition LFW AdaFace + WebFace4M + R100 Accuracy 0.9980 #7 of 16 Archive leaderboard report
Face Verification IJB-B AdaFace (WebFace4M) TAR@FAR=0.0001 96.03 #10 of 12 Archive leaderboard report
Face Verification IJB-B AdaFace (MS1MV3) TAR@FAR=0.0001 95.84 #11 of 12 Archive leaderboard report
Face Verification IJB-B AdaFace (MS1MV2) TAR@FAR=0.0001 95.67 #12 of 12 Archive leaderboard report
Face Verification IJB-C AdaFace (WebFace4M) TAR @ FAR=1e-4 97.39% #13 of 26 Archive leaderboard report
Face Verification IJB-C AdaFace (MS1MV3) TAR @ FAR=1e-4 97.09% #16 of 26 Archive leaderboard report
Face Verification IJB-C AdaFace (MS1MV2) TAR @ FAR=1e-4 96.89% #17 of 26 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