Learnable Privacy-Preserving Anonymization for Pedestrian Images

TL;DR

This paper introduces a reversible anonymization framework for pedestrian images that balances privacy preservation with the utility of re-identification, enabling effective privacy protection without sacrificing identification accuracy.

Contribution

It proposes a joint learning reversible anonymization method with a progressive training strategy, improving privacy-utility trade-off for pedestrian images.

Findings

Boosts re-identification performance with anonymized images

Maintains privacy by preventing third-party recognition

Reversibly generates full-body anonymous images

Abstract

This paper studies a novel privacy-preserving anonymization problem for pedestrian images, which preserves personal identity information (PII) for authorized models and prevents PII from being recognized by third parties. Conventional anonymization methods unavoidably cause semantic information loss, leading to limited data utility. Besides, existing learned anonymization techniques, while retaining various identity-irrelevant utilities, will change the pedestrian identity, and thus are unsuitable for training robust re-identification models. To explore the privacy-utility trade-off for pedestrian images, we propose a joint learning reversible anonymization framework, which can reversibly generate full-body anonymous images with little performance drop on person re-identification tasks. The core idea is that we adopt desensitized images generated by conventional methods as the initial privacy-preserving supervision and jointly train an anonymization encoder with a recovery decoder and an identity-invariant model. We further propose a progressive training strategy to improve the performance, which iteratively upgrades the initial anonymization supervision. Experiments further demonstrate the effectiveness of our anonymized pedestrian images for privacy protection, which boosts the re-identification performance while preserving privacy. Code is available at \url{https://github.com/whuzjw/privacy-reid}.