面向隐私保护可见光-红外行人重识别的跨模态保持、一致性与对齐
Towards Cross-Modal Preservation, Consistency and Alignment for Privacy-Preserving Visible-Infrared Person Re-Identification
针对可见光-红外行人重识别中的隐私保护问题，提出关键点保持正则化与差分一致性引导的模态对齐，可复用于跨模态检索中的匿名化特征对齐。
原文依据 we introduce a new task: Privacy-Preserving Visible-Infrared Person Re-Identification (PP-VI-ReID)
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论隐私保护行人重识别（PP-ReID）通过在多台非重叠摄像头间检索行人并应用匿名化技术保护敏感信息，解决了重识别中隐私与效用之间的核心权衡问题。然而，现有PP-ReID研究局限于单模态可见光场景，而24小时监控系统需要鲁棒的跨模态可见光-红外（VI）能力。因此，将PP-ReID扩展到跨模态VI场景至关重要。为此，我们引入一项新任务：隐私保护可见光-红外行人重识别（PP-VI-ReID）。该任务面临两大严峻挑战：1）粗糙的匿名化策略破坏身份关键信息并扰乱跨模态对齐；2）匿名化过程在不同模态间产生不一致的失真——在可见光图像中破坏基于颜色的纹理，而在红外图像中模糊热轮廓。这种不一致性与模态差距共同形成混合差距。为克服这些挑战，我们提出一个框架——精确隐私保护与对齐网络（PPA），包含两个组件：1）关键点保持正则化（KPR）模块利用人体姿态作为先验，引导结构感知的匿名化，保留关键身体特征；2）差分一致性引导的模态对齐（DCMA）将匿名化扰动视为稳定的、可学习的偏移而非变化的噪声，促进跨模态原始特征与匿名化特征之间的鲁棒对齐。在SYSU-MM01和RegDB上的实验验证了我们的框架，并建立了强基线。
英文原文摘要 1619 字符
Privacy-preserving Person Re-Identification (PP-ReID) addresses the core privacy-utility trade-off in Re-ID by retrieving a person across multiple non-overlapping cameras while applying anonymization techniques to protect sensitive information. However, prior PP-ReID studies are confined to single-modality visible scenarios, as 24-hour surveillance systems require robust cross-modal visible-infrared (VI) capabilities. Extending PP-ReID to the cross-modal VI setting is therefore crucial. Accordingly, we introduce a new task: Privacy-Preserving Visible-Infrared Person Re-Identification (PP-VI-ReID). This task presents two severe challenges: 1) Crude anonymization strategies destroy identity-critical information and disrupt cross-modal alignment 2) The anonymization process creates inconsistent distortions across modalities. It disrupts color-based textures in visible images while obscuring thermal contours in infrared images. This inconsistency with modality gap forms a Mixed Gap. To overcome these challenges, we propose a framework, the Precise Privacy-preserving and Alignment Network (PPA) with two components: 1) A Keypoint-Preserving Regularization (KPR) module leverages human pose as a prior to guide structure-aware anonymization, preserving essential body features. 2) A Differential Consistency-guided Modality Alignment (DCMA) treats anonymization perturbations not as varying noise, but as a stable, learnable offset, facilitating robust alignment between raw and anonymized features across modalities. Experiments on SYSU-MM01 and RegDB validate our framework, establishing a strong baseline.