标题： Pr$^2$R：终身人员再识别的信息融合与风格感知隐私保护重放 显示英文标题

标题： Pr$^2$R: Information-Fused and Style-Aware Privacy-Preserving Replay for Lifelong Person Re-Identification

摘要： 终身人物再识别（LReID）旨在在领域变化下，通过一系列任务逐步积累知识。 最近，基于重放的方法通过在辅助内存中复习过去样本，展示了在LReID中的强大有效性。 然而，存储历史示例引发了数据隐私方面的担忧。 为了避免这一点，无示例方法试图在不存储原始样本的情况下匹配过去数据的分布。 尽管这些方法具有隐私友好性，但由于特定过去知识表示的遗忘，它们通常会遭受性能下降。 为此，我们提出将顺序数据的信息融合到重放内存的像素空间中，从而实现隐私保护的重放（Pr$^2$R）。 具体来说，通过将多个真实图像的训练特征提炼到一张图像中，融合后的样本会经历像素级的变化。 这不仅保护了原始数据的隐私，还使重放样本对于顺序任务更具代表性。 在风格重放阶段，我们将当前领域与之前领域对齐，同时适应重放样本以匹配当前领域的风格。 这种双重对齐策略有效缓解了类别增量挑战和由领域变化引起的遗忘。 在多个基准上的广泛实验表明，所提出的方法在保持数据隐私的同时显著提高了重放效果。 具体而言，Pr$^2$R在顺序任务上的准确率分别比当前最先进的方法和其他基于重放的方法高出4%和6%。 显示英文摘要

摘要： Lifelong person re-identification (LReID) aims to incrementally accumulate knowledge across a sequence of tasks under domain shifts. Recently, replay-based methods have demonstrated strong effectiveness in LReID by rehearsing past samples stored in an auxiliary memory. However, storing historical exemplars raises concerns over data privacy. To avoid this, exemplar-free approaches attempt to match the distribution of past data without storing raw samples. Despite being privacy-friendly, these methods often suffer from performance degradation due to the forgetting of specific past knowledge representations. To this end, we propose to fuse information from sequential data into the pixel space in the replay memory, enabling Privacy-Preserving Replay (Pr$^2$R). More specifically, by distilling the training characteristics of multiple real images into a single image, the fused samples undergo pixel-level changes. This not only protects the privacy of the original data but also makes the replay samples more representative for sequential tasks. During the style replay phase, we align the current domain to the previous one while simultaneously adapting the replay samples to match the style of the current domain. This dual-alignment strategy effectively mitigates both class-incremental challenges and forgetting caused by domain shifts. Extensive experiments on multiple benchmarks show that the proposed method significantly improves replay effectiveness while preserving data privacy. Specifically, Pr$^2$R achieves 4% and 6% higher accuracy on sequential tasks compared to the current state-of-the-art and other replay-based methods, respectively.