用强化学习将图像转为隐私文本，兼顾安全与信息完整。
RL-MoE: An Image-Based Privacy Preserving Approach In Intelligent Transportation System
- 用专家混合模型分解场景，再由强化学习优化文本描述
- 在CFP-FP数据集上将重放攻击成功率降至9.4%
- 适合智能交通等需保护隐私的高敏感场景
智能交通系统中人工智能摄像头的普及，使视觉数据需求与隐私权之间产生严重冲突。现有隐私保护方法如模糊或加密，常因在隐私与数据效用间造成不可接受的权衡而效果不佳。为此，我们提出RL-MoE框架，将敏感视觉数据转换为隐私保护的文本描述，避免直接传输图像。该框架创新性地结合了用于多维度场景分解的专家混合（MoE）架构与强化学习（RL）代理，后者优化生成文本以同时满足语义准确性和隐私保护双重目标。大量实验表明，RL-MoE在隐私保护方面表现优异，使重放攻击成功率在CFP-FP数据集上降低至9.4%，同时生成的文本内容比基线方法更丰富。本工作为隐私敏感领域构建可信人工智能系统提供了实用且可扩展的解决方案，推动更安全的智慧城市与自动驾驶网络发展。
原文摘要 · Abstract (English)
The proliferation of AI-powered cameras in Intelligent Transportation Systems (ITS) creates a severe conflict between the need for rich visual data and the right to privacy. Existing privacy-preserving methods, such as blurring or encryption, are often insufficient due to creating an undesirable trade-off where either privacy is compromised against advanced reconstruction attacks or data utility is critically degraded. To resolve this challenge, we propose RL-MoE, a novel framework that transforms sensitive visual data into privacy-preserving textual descriptions, eliminating the need for direct image transmission. RL-MoE uniquely combines a Mixture-of-Experts (MoE) architecture for nuanced, multi-aspect scene decomposition with a Reinforcement Learning (RL) agent that optimizes the generated text for a dual objective of semantic accuracy and privacy preservation. Extensive experiments demonstrate that RL-MoE provides superior privacy protection, reducing the success rate of replay attacks to just 9.4\% on the CFP-FP dataset, while simultaneously generating richer textual content than baseline methods. Our work provides a practical and scalable solution for building trustworthy AI systems in privacy-sensitive domains, paving the way for more secure smart city and autonomous vehicle networks.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。