Abstract
Street-level imagery contains personally identifiable information (PII), some of which is context-dependent. Existing anonymization methods either over-process images or miss subtle identifiers, while API-based solutions compromise data sovereignty. We present an agentic framework CAIAMAR (\underline{C}ontext-\underline{A}ware \underline{I}mage \underline{A}nonymization with \underline{M}ulti-\underline{A}gent \underline{R}easoning) for context-aware PII segmentation with diffusion-based anonymization, combining pre-defined processing for high-confidence cases with multi-agent reasoning for indirect identifiers. Three specialized agents coordinate via round-robin speaker selection in a Plan-Do-Check-Act (PDCA) cycle, enabling large vision-language models to classify PII based on spatial context (private vs. public property) rather than rigid category rules. The agents implement spatially-filtered coarse-to-fine detection where a scout-and-zoom strategy identifies candidates, open-vocabulary segmentation processes localized crops, and $IoU$-based deduplication ($30\%$ threshold) prevents redundant processing. Modal-specific diffusion guidance with appearance decorrelation substantially reduces re-identification (Re-ID) risks. On CUHK03-NP, our method reduces person Re-ID risk by $73\%$ ($R1$: $16.9\%$ vs. $62.4\%$ baseline). For image quality preservation on CityScapes, we achieve KID: $0.001$, and FID: $9.1$, significantly outperforming existing anonymization. The agentic workflow detects non-direct PII instances across object categories, and downstream semantic segmentation is preserved. Operating entirely on-premise with open-source models, the framework generates human-interpretable audit trails supporting EU's GDPR transparency requirements while flagging failed cases for human review.
Abstract (translated)
街景图像包含个人身份信息（PII），其中部分信息的识别依赖于具体情境。现有匿名化方法要么对图像过度处理，要么遗漏细微标识，而基于API的方案则损害数据主权。本研究提出一种智能体框架CAIAMAR（上下文感知图像匿名化多智能体推理），通过基于扩散的匿名化实现上下文感知的PII分割，结合高置信度案例的预定义处理与针对间接标识的多智能体推理。三个专业化智能体通过PDCA循环中的轮询发言选择机制协同工作，使大视觉语言模型能够基于空间上下文（私有与公共财产）而非僵化的类别规则对PII进行分类。智能体实施空间过滤的粗到精检测策略：通过“侦察-放大”识别候选区域，开放词汇分割处理局部裁剪，并采用基于交并比（IoU）的去重（30%阈值）避免冗余处理。模态特定的扩散引导与外观去关联显著降低了重识别（Re-ID）风险。在CUHK03-NP数据集上，该方法将行人重识别风险降低73%（R1指标：16.9% vs. 基线62.4%）。在CityScapes数据集上，图像质量保持达到KID：0.001、FID：9.1，显著优于现有匿名化方法。该智能体工作流可跨对象类别检测非直接PII实例，且下游语义分割任务得以保留。框架完全基于本地部署并使用开源模型，生成人类可读的审计追踪以满足欧盟GDPR透明度要求，同时标记失败案例供人工复核。
URL
https://arxiv.org/abs/2603.27817
