Facial Expression Recognition with Controlled Privacy Preservation and Feature Compensation

Feng Xu1,2, David Ahmedt-Aristizabal2, Peterson Lars2, Dadong Wang2, Xun Li2,
1UNSW Sydney, 2CSIRO's Data61
Paper arXiv Code

The project website is under construction. Please check back later.

structure

Overview of the proposed method.

Abstract

Facial expression recognition (FER) systems raise significant privacy concerns due to the potential exposure of sensitive identity information. This paper presents a study on removing identity information while preserving FER capabilities.

Drawing on the observation that low-frequency components predominantly contain identity information and high-frequency components capture expression, we propose a novel two-stream framework that applies privacy enhancement to each component separately. We introduce a controlled privacy enhancement mechanism to optimize performance and a feature compensator to enhance task-relevant features without compromising privacy.

Furthermore, we propose a novel privacy-utility trade-off, providing a quantifiable measure of privacy preservation efficacy in closed-set FER tasks. Extensive experiments on the benchmark CREMA-D dataset demonstrate that our framework achieves 78.84% recognition accuracy with a privacy (facial identity) leakage ratio of only 2.01%, highlighting its potential for secure and reliable video-based FER applications.

BibTeX

We provide bibtex entries below: 


      @article{xu2024facial,
        title={Facial Expression Recognition with Controlled Privacy Preservation and Feature Compensation},
        author={Xu, Feng and Ahmedt-Aristizabal, David and Lars, Peterson and Wang, Dadong and Li, Xun},
        journal={arXiv preprint arXiv:2412.00277},
        year={2024}
      }