Papers › Killing Two Birds with One Stone:Efficient and Robust Training of Face Recognition...

Killing Two Birds with One Stone:Efficient and Robust Training of Face Recognition CNNs by Partial FC

28 Mar 2022arXiv:2203.15565archive 2025-07-28

Xiang An, Jiankang Deng, Jia Guo, Ziyong Feng, Xuhan Zhu, Jing Yang, Tongliang Liu

Learning discriminative deep feature embeddings by using million-scale in-the-wild datasets and margin-based softmax loss is the current state-of-the-art approach for face recognition. However, the memory and computing cost of the Fully Connected (FC) layer linearly scales up to the number of identities in the training set. Besides, the large-scale training data inevitably suffers from inter-class conflict and long-tailed distribution. In this paper, we propose a sparsely updating variant of the FC layer, named Partial FC (PFC). In each iteration, positive class centers and a random subset of negative class centers are selected to compute the margin-based softmax loss. All class centers are still maintained throughout the whole training process, but only a subset is selected and updated in each iteration. Therefore, the computing requirement, the probability of inter-class conflict, and the frequency of passive update on tail class centers, are dramatically reduced. Extensive experiments across different training data and backbones (e.g. CNN and ViT) confirm the effectiveness, robustness and efficiency of the proposed PFC. The source code is available at \https://github.com/deepinsight/insightface/tree/master/recognition.

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

No sample from the archive's official repository (deepinsight/insightface) is linked to this paper; the 12 samples below come from 3 community repositories. 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 12 code samples harvested from 3 repositories linked to this paper: 10 ran (of which 3 constructed an object rather than computing a result; 9 with no instrument failure: 1 honoured, 0 violated, 8 with no contract checked; 1 where Syntology's instrument failed) · 2 unverified. Where Syntology's instrument failed: 1 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: community (archive-listed): 12 samples from 3 repositories, 10 ran (9 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.

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 2026-09-26 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

deepinsight/insightface officialmentioned in papermentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). 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. report
bovifocr/insightface mentioned on GitHubpytorchSyntology: 1 sample, 0 ran · 1 unverified report
pedrobvidal/insightface mentioned on GitHubpytorchSyntology: 4 samples, 3 ran (0 instrument-failed) · 1 unverified report
yangyucheng000/-insightface mentioned on GitHubpytorchMITSyntology: 7 samples, 7 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.

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

12 samples harvested: 10 ran (of which 3 constructed an object rather than computing a result; 9 with no instrument failure: 1 honoured, 0 violated, 8 with no contract checked; 1 where Syntology's instrument failed) · 2 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

1ran · honoured contract
1ran · our draft was wrong
8ran
2unverified

Licence: 5 of the 12 samples are pointer only, meaning Syntology does not serve that copy's text. For 1 of them, this paper's copy is under a permissive licence and cleared, but the stored copy of the same code that Syntology serves is not; the licence label's tooltip gives Syntology's recorded reason. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 3 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. 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.

bovifocr/insightface community (archive-listed) 1 sample, 0 ran · 1 unverified
PartialFC recognition/arcface_torch/partial_fc.py unverified MIT (permissive) · ca2e74d15a48deb6 · report
pedrobvidal/insightface community (archive-listed) 4 samples, 3 ran (0 instrument-failed) · 1 unverified
AllGatherFunc recognition/arcface_torch/partial_fc_v2.py ran no licence file found · pointer only · 9599ba720223efe2 · report fixture: returned AllGatherFunc · level recorded 2026-09-22
DistCrossEntropy recognition/arcface_torch/partial_fc_v2.py ran · metamorphic tier: deterministic no licence file found · pointer only · 4f589c1b7b15107e · report fixture: returned DistCrossEntropy · level recorded 2026-09-22
DistCrossEntropyFunc recognition/arcface_torch/partial_fc_v2.py ran no licence file found · pointer only · c0df452dc531921a · report fixture: returned DistCrossEntropyFunc · level recorded 2026-09-22
PartialFC_V2 recognition/arcface_torch/partial_fc_v2.py unverified no licence file found · pointer only · 3a6ac854256554fc · report
yangyucheng000/-insightface community (archive-listed) 7 samples, 7 ran (1 instrument-failed) · 0 unverified
distance2bbox python-package/insightface/model_zoo/retinaface.py ran MIT (permissive) · 1d23600a36442d09 · report fixture: signature points:Tensor/2/float, distance:Tensor/2/float, max_shape:tuple/1/int · returned ndarray · level recorded 2026-09-24
distance2kps python-package/insightface/model_zoo/retinaface.py ran MIT (permissive) · e8f5d8d07e1e560b · report fixture: signature points:Tensor/2/float, distance:Tensor/2/float, max_shape:tuple/1/int · returned ndarray · level recorded 2026-09-24
find_onnx_file python-package/insightface/model_zoo/model_zoo.py ran fingerprinted MIT (permissive) · fc633d5737805a00 · report fixture: signature dir_path:str/0/str · returned str · level recorded 2026-09-24
find_params_file python-package/insightface/model_zoo/model_store.py ran fingerprinted MIT (permissive) · 8e01f2544321014d · report fixture: signature dir_path:str/0/str · returned str · level recorded 2026-09-24
get_model_file in 15 papers python-package/insightface/model_zoo/model_store.py ran MIT (permissive) · 36d98138ea733ae9 · report fixture: returned str · level recorded 2026-09-24
short_hash in 15 papers python-package/insightface/model_zoo/model_store.py ran · our draft was wrong MIT (permissive) here; the stored copy is pointer only · 48aa1fdc7e6172f9 · report fixture: signature name:str/0/str · returned str · level recorded 2026-09-26
softmax in 240 papers
census: 53 implementations from 63 papers, 16 distinct outputs across 8 buckets, one shared input per bucket
python-package/insightface/model_zoo/retinaface.py ran · honoured contract fingerprinted MIT (permissive) · 011537d0bea7fe09 · report fixture: signature z:ndarray/2/float · returned ndarray · level recorded 2026-09-24

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 Partial FC (PFC) (harvested), lifted in the same batch on 2026-09-09: alternative readings kept side by side; none replaces or supersedes another.

harvested Assembly for Partial FC (PFC) (graph method: Partial FC (PFC)). 4 bricks; individually on a fixture: 3 ran (of which 3 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified; every one of the 3 samples that ran constructed an object rather than computing a result. The assembly has not been run end-to-end.

one of 2 siblings

  1. PartialFC_V2 entry · level 0 · unverified · 3a6ac854256554fc
  2. DistCrossEntropy helper · level 1 · ran · 4f589c1b7b15107e
  3. DistCrossEntropyFunc loss · level 1 · ran · c0df452dc531921a
  4. AllGatherFunc helper · level 1 · ran · 9599ba720223efe2

harvested Assembly for Partial FC (PFC) (graph method: Partial FC (PFC)). 4 bricks; individually on a fixture: 3 ran (of which 3 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified; every one of the 3 samples that ran constructed an object rather than computing a result. The assembly has not been run end-to-end.

one of 2 siblings

  1. PartialFC entry · level 0 · unverified · ca2e74d15a48deb6
  2. DistCrossEntropy loss · level 1 · ran · 4f589c1b7b15107e
  3. DistCrossEntropyFunc loss · level 1 · ran · c0df452dc531921a
  4. AllGatherFunc helper · level 1 · ran · 9599ba720223efe2

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="2203.15565") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Face RecognitionFace Verification

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Face Recognition MFR Partial FC African 98.07 #1 of 1 Archive leaderboard report
Face Recognition MFR Partial FC Caucasian 98.81 #1 of 1 Archive leaderboard report
Face Recognition MFR Partial FC East Asian 89.97 #1 of 1 Archive leaderboard report
Face Recognition MFR Partial FC MFR-ALL 97.85 #1 of 1 Archive leaderboard report
Face Recognition MFR Partial FC MFR-MASK 90.88 #1 of 1 Archive leaderboard report
Face Recognition MFR Partial FC South Asian 98.66 #1 of 1 Archive leaderboard report
Face Verification AgeDB-30 PartialFC(R200) Accuracy 0.9870 #1 of 5 Archive leaderboard report
Face Verification CFP-FP PartialFC (R200) Accuracy 0.9951 #1 of 4 Archive leaderboard report
Face Verification IJB-B PartialFC(WebFace42M) TAR@FAR=0.0001 96.71 #9 of 12 Archive leaderboard report
Face Verification IJB-C Partial FC TAR @ FAR=1e-4 98.00% #4 of 26 Archive leaderboard report
Face Verification IJB-C Partial FC TAR @ FAR=1e-5 97.23% #4 of 26 Archive leaderboard report
Face Verification IJB-C Partial FC model ViT-L #4 of 26 Archive leaderboard report
Face Verification IJB-C Partial FC training dataset WebFace42M #4 of 26 Archive leaderboard report
Face Verification IJB-C PartialFC TAR @ FAR=1e-4 97.97% #5 of 26 Archive leaderboard report
Face Verification IJB-C PartialFC TAR @ FAR=1e-5 96.93% #5 of 26 Archive leaderboard report
Face Verification IJB-C PartialFC model R200 #5 of 26 Archive leaderboard report
Face Verification IJB-C PartialFC training dataset WebFace42M #5 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.

Methods

Softmax

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