AI Paper Atlas

Research map / Benchmarks / LFW

LFW leaderboard

91 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on LFW. Best accuracy: 99.86 (LAFS, CVPR 2024).

Numbers are as each paper reports them, read by AI from its results tables and not checked by hand. Papers can differ in setting (training data, model size, extra data); the topic column shows what each paper is about. Open the paper before you compare or quote a number.

LFW · accuracy

38 papers, each with its best reported number for its own method; higher is better.

28 more papers on this leaderboard, the baselines reported on it and a CSV download are free with a scime account.
Sign in free to see all

LFW · verification accuracy (%)

37 papers, each with its best reported number for its own method; higher is better.

#verification accuracy (%) ↑MethodPaperTopic
199.85GIF
R100
GIF: Generative Inspiration for Face Recognition at Scale
CVPR 2025 · Table 1
Recognition and long-tail
299.853D GAN (FFHQ), 20k synth IDs
ResNet-100
A 3D GAN for Improved Large-Pose Facial Recognition
CVPR 2021 · Table 2
Face and shape
399.85GroupFace
ResNet-100
GroupFace: Learning Latent Groups and Constructing Group-Based Representations for Face Recognition
CVPR 2020 · Table: Verification accuracy o
Recognition and long-tail
499.85BroadFaceBroadFace: Looking at Tens of Thousands of People at Once for Face Recognition
ECCV 2020 · Table 1
Recognition and long-tail
599.833Prodpoly-ResNet50
ResNet50
Deep Polynomial Neural Networks
TPAMI 2021 · Table 9
RNNs and normalization
699.83A18+MSML
ArcFace-18
MSML: Enhancing Occlusion-Robustness by Multi-Scale Segmentation-Based Mask Learning for Face Recognition
AAAI 2022 · Table 1
Recognition and long-tail
799.83MagFace
ResNet100
MagFace: A Universal Representation for Face Recognition and Quality Assessment
CVPR 2021 · Table 1
Recognition and long-tail
899.83VPL-ArcFace
ResNet100
Variational Prototype Learning for Deep Face Recognition
CVPR 2021 · Table 6
Recognition and long-tail
999.83ArcFace trained on MovieFace
R-100 (ArcFace)
Caption-Supervised Face Recognition: Training a State-of-the-Art Face Model without Manual Annotation
ECCV 2020 · Table 2
Recognition and long-tail
1099.83ArcFace
MS1MV2, ResNet100
ArcFace: Additive Angular Margin Loss for Deep Face Recognition
CVPR 2019 · Table 4
Recognition and long-tail
27 more papers on this leaderboard, the baselines reported on it and a CSV download are free with a scime account.
Sign in free to see all

LFW · FID

16 papers, each with its best reported number for its own method; lower is better.

6 more papers on this leaderboard, the baselines reported on it and a CSV download are free with a scime account.
Sign in free to see all

Not affiliated with any conference, journal or the benchmark's authors. Corrections: contact@latenticlabs.com.