AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

WebFace260M: A Benchmark for Million-Scale Deep Face Recognition

Zheng Dong Zhu, Guan Huang, Jiankang Deng, Yun Ye, Junjie Huang, Xinze Chen, Jiagang Zhu, Tian Yang, Dalong Du, Jiwen Lu, Jie Zhou

TPAMI 2022 · Label-efficient and robust learning › Recognition and long-tail · approach: Fair and quality-aware face recognition · 68 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
train on WebFace42M instead of MS1MV2ArcFace ResNet-100 · IJB-CTAR@FAR=1e-4: 96.03 -> 97.70Improved
train on WebFace42M instead of MS1MV2ArcFace ResNet-100 · Our test set SFR AllFNMR@FMR=1e-5: 9.88 -> 2.98Improved
train on WebFace42M instead of MS1MV2CurricularFace ResNet-100 · Our test set MFR All-MaskedFNMR@FMR=1e-5: 69.56 -> 42.97Improved
increase WebFace portion from 10% to 30% to 100% (ablation)ArcFace ResNet-100 · IJB-CTAR@FAR=1e-4: 96.77 -> 97.47 -> 97.70Improved
increase WebFace portion from 10% to 100% (ablation)ArcFace ResNet-14 · IJB-CTAR@FAR=1e-4: 93.13 -> 93.92 -> 94.22Improved
add simulated mask augmentation to training data (ablation)ArcFace ResNet-100 on WebFace42M · Our test set MFR All-MaskedFNMR@FMR=1e-5: 47.25 -> 33.87Improved
add simulated mask augmentation to training data (ablation)ArcFace ResNet-100 on WebFace4M · Our test set MFR All-MaskedFNMR@FMR=1e-5: 70.97 -> 59.89Improved
add more CAST iterations (1 to 3) (ablation)ArcFace ResNet-100 on WebFace · IJB-CTAR@FAR=1e-4: 97.28 -> 97.51 -> 97.70Improved
add a 4th CAST iteration (ablation)ArcFace ResNet-100 on WebFace · IJB-CTAR@FAR=1e-4: 97.70 -> 97.66No clear effect
use ResNet-200 instead of ResNet-100 as CAST teacher/student (ablation)CAST · IJB-CTAR@FAR=1e-4: 97.70 -> 97.71No clear effect
use Glint360K instead of MS1MV2 for initial teacher (ablation)CAST · IJB-CTAR@FAR=1e-4: 97.70 -> 97.68No clear effect
use MobileFaceNet instead of ResNet-100 as CAST teacher/student (ablation)CAST · IJB-CTAR@FAR=1e-4: 97.70 -> 95.82Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy