Papers › ArcFace: Additive Angular Margin Loss for Deep Face Recognition
ArcFace: Additive Angular Margin Loss for Deep Face Recognition
Jiankang Deng, Jia Guo, Jing Yang, Niannan Xue, Irene Kotsia, Stefanos Zafeiriou
Recently, a popular line of research in face recognition is adopting margins in the well-established softmax loss function to maximize class separability. In this paper, we first introduce an Additive Angular Margin Loss (ArcFace), which not only has a clear geometric interpretation but also significantly enhances the discriminative power. Since ArcFace is susceptible to the massive label noise, we further propose sub-center ArcFace, in which each class contains K sub-centers and training samples only need to be close to any of the K positive sub-centers. Sub-center ArcFace encourages one dominant sub-class that contains the majority of clean faces and non-dominant sub-classes that include hard or noisy faces. Based on this self-propelled isolation, we boost the performance through automatically purifying raw web faces under massive real-world noise. Besides discriminative feature embedding, we also explore the inverse problem, mapping feature vectors to face images. Without training any additional generator or discriminator, the pre-trained ArcFace model can generate identity-preserved face images for both subjects inside and outside the training data only by using the network gradient and Batch Normalization (BN) priors. Extensive experiments demonstrate that ArcFace can enhance the discriminative feature embedding as well as strengthen the generative face synthesis.
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):
get_harvested_code_for_paper(arxiv_id="1801.07698")get_code(code_sha256="…")with a sample'scode_sha256from the list belowget_citation_path(paper_1="1801.07698", paper_2="…")with another paper's arXiv id or titleget_concepts_for_paper(arxiv_id="1801.07698")
Code
No sample from the archive's official repository (deepinsight/insightface) is linked to this paper; the 19 samples below come from 11 community repositories, and 2 more are identical to code Syntology harvested from another paper's repository. Syntology harvested code from deepinsight/insightface for another paper.
deepinsight/insightface: Syntology's graph links these samples to A Multimodal Dataset of Student Oral Presentations with Sensors and Evaluation Data (arXiv:2601.07576; that paper's own run record: 14 ran (of which 0 constructed an object rather than computing a result; 6 with no instrument failure: 3 honoured, 1 violated, 2 with no contract checked; 8 where Syntology's instrument failed) · 7 unverified); nothing checks that they implement this paper's method.
Syntology 21 code samples: 19 from 11 repositories linked to this paper, 2 identical to code harvested from another paper's repository. Across all 21: 16 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 2 honoured, 0 violated, 0 with no contract checked; 14 where Syntology's instrument failed) · 5 unverified. Where Syntology's instrument failed: 7 where our contract draft was wrong; 7 where the contract check could not run; failures of Syntology's instrument, not of the code.
By repository: community (archive-listed): 19 samples from 11 repositories, 15 ran (2 with no instrument failure; 13 where Syntology's instrument failed); 2 identical to code first harvested elsewhere. 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 21 of its repositories that Syntology harvested for other papers:
- deepinsight/insightface, official repository: harvested for another paper, listed under the repository
- Hayashi-Yudai/ML_models, community (archive-listed): harvested for another paper, listed under the repository
- IrvingMeng/MagFace, community (archive-listed): harvested for another paper, listed under the repository
- PaddlePaddle/PLSC, community (archive-listed): harvested for another paper, listed under the repository
- SevenZhan/Pytorch, community (archive-listed): harvested for another paper, listed under the repository
- TreB1eN/InsightFace_Pytorch, community (archive-listed): harvested for another paper, listed under the repository
- code-trip/insightface-mxnet, community (archive-listed): harvested for another paper, listed under the repository
- godofpdog/MobileFaceNet_keras, community (archive-listed): harvested for another paper, listed under the repository
- nhatduy19599/insightface-master, community (archive-listed): harvested for another paper, listed under the repository
- peteryuX/arcface-tf2, community (archive-listed): harvested for another paper, listed under the repository
- samisoto/keras_cosine_based_loss, community (archive-listed): harvested for another paper, listed under the repository
- sunil-rival/insightface, community (archive-listed): harvested for another paper, listed under the repository
- trachpro/arcface-tf2, community (archive-listed): harvested for another paper, listed under the repository
- vladimirwest/insightface_cinematic, community (archive-listed): harvested for another paper, listed under the repository
- vvictoryuki/freedom, community (archive-listed): harvested for another paper, listed under the repository
- 2024-MindSpore-1/Code4, community (archive-listed): harvested for another paper, listed under the repository
- Armxyz1/Results-on-RFW, community (archive-listed): harvested for another paper, listed under the repository
- RocketFlash/easy_metric_learning, community (archive-listed): harvested for another paper, listed under the repository
- mindspore-ai/models, community (archive-listed): harvested for another paper, listed under the repository
- sajjjadayobi/FaceLib, community (archive-listed): harvested for another paper, listed under the repository
- serengil/deepface, community (archive-listed): harvested for another paper, listed under the repository
Levels written on 2026-09-22 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed. 2 of the 16 samples that ran have no recorded date.
100 repositories listed; official and paper-mentioned ones first.
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
21 samples harvested: 16 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 2 honoured, 0 violated, 0 with no contract checked; 14 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.
Licence: 14 of the 21 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 11 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. 4 samples whose link to this paper records no repository are listed under one of this paper's own repositories, because identical code was harvested from it for another paper. 2 samples are listed last, in a group without a repository. “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.
5b24fc008e2c830a · report
eddb4fc2f9316aa0 · report
fixture: signature arg_params:dict/0/?, aux_params:dict/0/?, ctx:MockContext/0/? · returned tuple · level recorded 2026-09-22 fac5364e2f53c6db · report
fixture: signature in_planes:int/0/int, out_planes:int/0/int, stride:int/0/int · returned Conv2d · level recorded, date not recorded e984e4da60619384 · report
64b45da43b8ab573 · report
fixture: signature thresholds:ndarray/1/float, embeddings1:ndarray/2/float, embeddings2:ndarray/2/float, actual_issame:ndarray/1/bool, nrof_folds:int/0/int, pca:int/0/int · returned tuple · level recorded 2026-09-22 f67f5f081381cc8a · report
fixture: signature thresholds:ndarray/1/float, embeddings1:ndarray/2/float, embeddings2:ndarray/2/float, actual_issame:ndarray/1/bool, far_target:float/0/float, nrof_folds:int/0/int · returned tuple · level recorded 2026-09-22 census: 8 implementations from 7 papers, 7 distinct outputs across 6 buckets, one shared input per bucket lfw_eval.py unverified Apache-2.0 (permissive) ·
dae42d7f4d41b99b · report
5f529c4af000c070 · report
fixture: signature thresholds:ndarray/1/float, embeddings1:ndarray/2/float, embeddings2:ndarray/2/float, actual_issame:ndarray/1/bool, nrof_folds:int/0/int, pca:int/0/int · returned tuple · level recorded 2026-09-22 c2ce20a5d37df6c4 · report
fixture: signature threshold:float/0/float, dist:ndarray/1/float, actual_issame:ndarray/1/bool · returned tuple · level recorded 2026-09-22 9348919b3a8be935 · report
fixture: signature Reframe:list/1/float, GTframe:list/1/float · returned float · level recorded 2026-09-22 c2a17396e5b1b237 · report
fixture: signature argv:list/1/str · returned Namespace · level recorded 2026-09-22 29507ad0ee182d8c · report
fixture: signature img:ndarray/2/int · returned ndarray · level recorded 2026-09-22 151ad8cf4f442cf0 · report
fixture: signature path:str/0/str · returned tuple · level recorded 2026-09-22 f975336b5d5f1a80 · report
fixture: signature threshold:float/0/float, dist:ndarray/1/float, actual_issame:ndarray/1/bool · returned tuple · level recorded 2026-09-22 db4008738f6c3eae · report
fixture: signature thresholds:ndarray/1/float, embeddings1:ndarray/2/float, embeddings2:ndarray/2/float, actual_issame:ndarray/1/bool, nrof_folds:int/0/int, pca:int/0/int · returned tuple · level recorded 2026-09-22 c7c89ea2508beb10 · report
fixture: signature thresholds:ndarray/1/float, embeddings1:ndarray/2/float, embeddings2:ndarray/2/float, actual_issame:ndarray/1/bool, far_target:float/0/float, nrof_folds:int/0/int · returned tuple · level recorded 2026-09-22 bd8fb5c95fa87652 · report
fixture: signature pretrained:bool/0/bool · returned ResNet · level recorded 2026-09-22 bb357f89cac141ff · report
fixture: signature pretrained:bool/0/bool · returned ResNet · level recorded 2026-09-22 2 samples are identical to code Syntology harvested from another paper's repository; this paper's own copy was not located, so no repository is named and no licence is recorded.
2da848988b60d9cc · report
fixture: signature listTemp:list/1/int, n:int/0/int · returned list · level recorded, date not recorded 897480715a729925 · report
Tasks
Results from the paper archive 2025-07-28
| Task | Dataset | Model | Metric | Value | Rank at snapshot | Leaderboard | Report |
|---|---|---|---|---|---|---|---|
| Face Identification | MegaFace | ArcFace + MS1MV2 + R100 + R | Accuracy | 98.35% | #6 of 13 | Archive leaderboard | report |
| Face Identification | Trillion Pairs Dataset | Arc-Softmax | Accuracy | 57.48 | #3 of 6 | Archive leaderboard | report |
| Face Recognition | CASIA-WebFace+masks | ArcFace | Accuracy | 87.95 | #3 of 6 | Archive leaderboard | report |
| Face Recognition | CelebA+masks | ArcFace | Accuracy | 91.78 | #3 of 6 | Archive leaderboard | report |
| Face Verification | IJB-C | ArcFace | TAR @ FAR=1e-5 | 96.07% | #6 of 26 | Archive leaderboard | report |
| Face Verification | IJB-C | ArcFace | model | R100 | #6 of 26 | Archive leaderboard | report |
| Face Verification | IJB-C | ArcFace | training dataset | IBUG-500K | #6 of 26 | Archive leaderboard | report |
| Face Verification | Labeled Faces in the Wild | ArcFace + MS1MV2 + R100, | Accuracy | 99.83% | #1 of 7 | Archive leaderboard | report |
| Face Verification | MegaFace | ArcFace + MS1MV2 + R100 + R | Accuracy | 98.48% | #4 of 12 | Archive leaderboard | report |
| Face Verification | Trillion Pairs Dataset | Arc-Softmax | Accuracy | 57.45 | #3 of 6 | Archive leaderboard | report |
| Face Verification | YouTube Faces DB | ArcFace + MS1MV2 + R100, | Accuracy | 98.02% | #2 of 12 | 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
Introduced by this paper: ArcFace
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