Person Re-Identification without Identification via Event Anonymization

TL;DR

This paper introduces a novel method to anonymize event-streams from neuromorphic sensors, protecting individual privacy while enabling person re-identification, and presents the first event-based person ReId dataset for evaluation.

Contribution

It proposes an end-to-end network that scrambles event data to prevent image reconstruction attacks while maintaining re-identification performance, and provides a new dataset for this task.

Findings

Effective anonymization of event streams demonstrated

Preserved person ReId accuracy despite scrambling

First dataset for event-based person ReId introduced

Abstract

Wide-scale use of visual surveillance in public spaces puts individual privacy at stake while increasing resource consumption (energy, bandwidth, and computation). Neuromorphic vision sensors (event-cameras) have been recently considered a valid solution to the privacy issue because they do not capture detailed RGB visual information of the subjects in the scene. However, recent deep learning architectures have been able to reconstruct images from event cameras with high fidelity, reintroducing a potential threat to privacy for event-based vision applications. In this paper, we aim to anonymize event-streams to protect the identity of human subjects against such image reconstruction attacks. To achieve this, we propose an end-to-end network architecture jointly optimized for the twofold objective of preserving privacy and performing a downstream task such as person ReId. Our network learns to scramble events, enforcing the degradation of images recovered from the privacy attacker. In this work, we also bring to the community the first ever event-based person ReId dataset gathered to evaluate the performance of our approach. We validate our approach with extensive experiments and report results on the synthetic event data simulated from the publicly available SoftBio dataset and our proposed Event-ReId dataset.