Federated Face Forgery Detection Learning with Personalized Representation

TL;DR

This paper introduces a personalized federated learning approach for face forgery detection, enhancing model performance while preserving privacy across distributed client devices, and demonstrates superior results on public datasets.

Contribution

It proposes a novel federated learning framework with personalized representations specifically designed for face forgery detection, addressing generalization issues in non-public data scenarios.

Findings

Outperforms state-of-the-art methods on public datasets

Improves detection accuracy through personalized representations

Enhances privacy preservation in distributed learning environments

Abstract

Deep generator technology can produce high-quality fake videos that are indistinguishable, posing a serious social threat. Traditional forgery detection methods directly centralized training on data and lacked consideration of information sharing in non-public video data scenarios and data privacy. Naturally, the federated learning strategy can be applied for privacy protection, which aggregates model parameters of clients but not original data. However, simple federated learning can't achieve satisfactory performance because of poor generalization capabilities for the real hybrid-domain forgery dataset. To solve the problem, the paper proposes a novel federated face forgery detection learning with personalized representation. The designed Personalized Forgery Representation Learning aims to learn the personalized representation of each client to improve the detection performance of individual client models. In addition, a personalized federated learning training strategy is utilized to update the parameters of the distributed detection model. Here collaborative training is conducted on multiple distributed client devices, and shared representations of these client models are uploaded to the server side for aggregation. Experiments on several public face forgery detection datasets demonstrate the superior performance of the proposed algorithm compared with state-of-the-art methods. The code is available at \emph{https://github.com/GANG370/PFR-Forgery.}