A Dual-stage Prompt-driven Privacy-preserving Paradigm for Person Re-Identification

TL;DR

This paper introduces a dual-stage prompt-driven framework for privacy-preserving person re-identification, generating a large virtual dataset and learning domain-invariant features to improve model generalization in real-world scenarios.

Contribution

The paper proposes a novel dual-stage prompt-driven paradigm with a new dataset and a disentanglement mechanism to enhance privacy and domain generalization in person Re-ID.

Findings

Models trained on GenePerson with PDM outperform existing datasets.

The approach achieves state-of-the-art generalization performance.

Synthetic data effectively replaces real data for privacy-preserving Re-ID.

Abstract

With growing concerns over data privacy, researchers have started using virtual data as an alternative to sensitive real-world images for training person re-identification (Re-ID) models. However, existing virtual datasets produced by game engines still face challenges such as complex construction and poor domain generalization, making them difficult to apply in real scenarios. To address these challenges, we propose a Dual-stage Prompt-driven Privacy-preserving Paradigm (DPPP). In the first stage, we generate rich prompts incorporating multi-dimensional attributes such as pedestrian appearance, illumination, and viewpoint that drive the diffusion model to synthesize diverse data end-to-end, building a large-scale virtual dataset named GenePerson with 130,519 images of 6,641 identities. In the second stage, we propose a Prompt-driven Disentanglement Mechanism (PDM) to learn domain-invariant generalization features. With the aid of contrastive learning, we employ two textual inversion networks to map images into pseudo-words representing style and content, respectively, thereby constructing style-disentangled content prompts to guide the model in learning domain-invariant content features at the image level. Experiments demonstrate that models trained on GenePerson with PDM achieve state-of-the-art generalization performance, surpassing those on popular real and virtual Re-ID datasets.