通过身份解耦和几何保留的人脸蒸馏发布私密人脸识别训练数据集
Private Face Recognition Training Dataset Publication via Identity-Decoupled and Geometry-Preserving Face Distillation
浏览论文内容
中文总结 AI 辅助
该研究针对私密人脸识别训练数据集发布的身份悖论，提出Private Face Distillation框架，通过身份解耦和几何保留实现隐私保护，在IJB-C监控任务上提升识别性能并降低源身份可关联性。
中文摘要 AI 辅助
发布私密人脸识别（FR）训练数据集具有隐私敏感性，因为人脸会暴露身份信息。私密FR训练数据集发布通过发布受保护的代理作为替代，替代私密训练人脸，来缓解这种风险。然而，使用此类数据训练FR模型会引发身份悖论：使发布的人脸对识别监督有用的身份线索，同时也是使它们可关联到真实个体的线索。受保护的人脸应与原始身份解耦，但仍需作为可靠的身份样本用于训练。过度移除这些线索可能会破坏识别学习所需的类别结构，而过度保留这些线索则会增加源身份的可关联性。我们认为，这一悖论源于将源对齐的身份语义与对识别有用的代理身份几何混为一谈：前者应被抑制以减少与私密个体的关联，而后者应被保留用于FR学习。基于这一见解，我们提出了Private Face Distillation，这是一个身份解耦和几何保留的框架。它使用正交几何保留从私密身份表示中构建解耦的代理身份，同时保持超球面几何，并使用关系拓扑对齐来保留用于识别学习的身份关系。在多个域偏移的FR场景中进行的实验表明，Private Face Distillation比所评估的发布基线实现了更强的效用。在IJB-C监控任务上，它在FAR=1e-3时的TAR比基线提高了3.94%，同时降低了源身份的可关联性。这些结果表明，私密FR训练数据集发布应在保留代理身份几何的同时，解耦源身份对应关系。
英文摘要
Publishing private face recognition~(FR) training datasets is privacy-sensitive because faces expose identity information. Private FR training dataset publication mitigates this risk by releasing protected proxies as substitutes for private training faces. However, training FR models with such data introduces an identity paradox: \emph{the identity cues that make released faces useful for recognition supervision are also the cues that make them linkable to real individuals.} A protected face should be decoupled from the original identity, yet still behave as a reliable identity sample for training. Removing these cues too aggressively may destroy the class structure needed for recognition learning, whereas preserving them too faithfully may increase source-identity linkability. We argue that this paradox stems from conflating source-aligned identity semantics with recognition-useful proxy identity geometry. The former should be suppressed to reduce linkage to private individuals, while the latter should be preserved for FR learning. Based on this insight, we propose \textbf{Private Face Distillation}, an identity-decoupling and geometry-preserving framework. It uses Orthogonal Geometry Preservation to construct decoupled proxy identities from private identity representations while maintaining hyperspherical geometry, and Relational Topology Alignment to preserve identity relations for recognition learning. Experiments across multiple domain-shifted FR scenarios show that Private Face Distillation achieves stronger utility than the evaluated publication baselines. On IJB-C surveillance, it improves $\mathrm{TAR}@\mathrm{FAR}{=}1\text{e-}{3}$ by 3.94\% over the baseline while reducing source-identity linkability. These results suggest that private FR training dataset publication should decouple source-identity correspondence while preserving proxy identity geometry.
