Federated Learning-Based Face Recognition: Methods, Challenges and Future Prospects

Xiaoying Yang

2024

Abstract

With the rapid development of face recognition technology, federated learning has become a widely used method for face recognition due to its distributed collaboration and privacy-preserving properties. This paper systematically introduces the existing work on federated learning for face recognition to provide a referenceable overview for research in this area. In order to understand the function of federated learning in different application scenarios of face recognition, this paper discusses the implementation of different models in detail, dissects the representative models of federated learning in solving the three aspects of privacy-preserving improvement, gradient correction, and small-sample image recognition, and sorts out and explains the working principle of the models to elucidate the advantages of them in applications. Then, the current challenges of federated learning for face recognition are presented, pointing out that the current issues of data heterogeneity, applicability expansion, and interpretability still need to be further researched and improved, and possible solutions for the future are proposed.

Paper Citation

in Harvard Style

Yang X. (2024). Federated Learning-Based Face Recognition: Methods, Challenges and Future Prospects. In Proceedings of the 1st International Conference on Engineering Management, Information Technology and Intelligence - Volume 1: EMITI; ISBN 978-989-758-713-9, SciTePress, pages 569-573. DOI: 10.5220/0012959400004508

in Bibtex Style

@conference{emiti24,
author={Xiaoying Yang},
title={Federated Learning-Based Face Recognition: Methods, Challenges and Future Prospects},
booktitle={Proceedings of the 1st International Conference on Engineering Management, Information Technology and Intelligence - Volume 1: EMITI},
year={2024},
pages={569-573},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012959400004508},
isbn={978-989-758-713-9},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on Engineering Management, Information Technology and Intelligence - Volume 1: EMITI
TI - Federated Learning-Based Face Recognition: Methods, Challenges and Future Prospects
SN - 978-989-758-713-9
AU - Yang X.
PY - 2024
SP - 569
EP - 573
DO - 10.5220/0012959400004508
PB - SciTePress