mlsanityData to 7 Oct 2026

Towards Large-scale Masked Face Recognition

arXiv 2310.16364 ↗First posted 25 Oct 2023cs.CVcited by 1

Not on the map. Why?

Abstract (arXiv)

During the COVID-19 coronavirus epidemic, almost everyone is wearing masks, which poses a huge challenge for deep learning-based face recognition algorithms. In this paper, we will present our championship solutions in ICCV MFR WebFace260M and InsightFace unconstrained tracks. We will focus on four challenges in large-scale masked face recognition, i.e., super-large scale training, data noise handling, masked and non-masked face recognition accuracy balancing, and how to design inference-friendly model architecture. We hope that the discussion on these four aspects can guide future research towards more robust masked face recognition systems.

Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 25 of its 35 references that are in the map. First posted on arXiv 25 Oct 2023.

Builds on

25 of 35 references.

References · 25

  1. 2014Microsoft COCO: Common Objects in Contextcited by 17,498
  2. 2019EfficientNet: Rethinking Model Scaling for Convolutional Neural Networkscited by 6,415
  3. 2017Mixed Precision Trainingcited by 1,135
  4. 2018CosFace: Large Margin Cosine Loss for Deep Face Recognitioncited by 968
  5. 2018MnasNet: Platform-Aware Neural Architecture Search for Mobilecited by 937
  6. 2017VGGFace2: A dataset for recognising faces across pose and agecited by 911
  7. 2017SphereFace: Deep Hypersphere Embedding for Face Recognitioncited by 733
  8. 2016MS-Celeb-1M: A Dataset and Benchmark for Large-Scale Face Recognitioncited by 621
  9. 2014Learning Face Representation from Scratchcited by 451
  10. 2021WebFace260M: A Benchmark Unveiling the Power of Million-Scale Deep Face Recognitioncited by 221
  11. 2015The MegaFace Benchmark: 1 Million Faces for Recognition at Scalecited by 199
  12. 2018Joint 3D Face Reconstruction and Dense Alignment with Position Map Regression Networkcited by 195
13 more
  1. 2020Partial FC: Training 10 Million Identities on a Single Machinecited by 163
  2. 2023FlowFormer++: Masked Cost Volume Autoencoding for Pretraining Optical Flow Estimationcited by 114
  3. 2023VideoFlow: Exploiting Temporal Cues for Multi-frame Optical Flow Estimationcited by 82
  4. 2018The Devil of Face Recognition is in the Noisecited by 65
  5. 2021FaceX-Zoo: A PyTorch Toolbox for Face Recognitioncited by 49
  6. 2021Masked Face Recognition Challenge: The InsightFace Track Reportcited by 41
  7. 2021Efficient Masked Face Recognition Method during the COVID-19 Pandemiccited by 15
  8. 2021Masked Face Recognition Challenge: The WebFace260M Track Reportcited by 12
  9. 2019Towards Flops-constrained Face Recognitioncited by 9
  10. 2023Decoupled DETR: Spatially Disentangling Localization and Classification for Improved End-to-End Object Detectioncited by 8
  11. 2022Towards Robust Face Recognition with Comprehensive Searchcited by 3
  12. 20201st place solution for AVA-Kinetics Crossover in AcitivityNet Challenge 2020cited by 2
  13. 2020Top-1 Solution of Multi-Moments in Time Challenge 2019cited by 1

Built on by

Cited by year

Citing papers per year: 2025: 1.

1 paper.

  1. 2025High-Fidelity Diffusion Face Swapping with ID-Constrained Facial Conditioningcited by 5