RoboShape：面向隐私感知机器人感知的信息论点云表示
RoboShape: Information-Theoretic Point Cloud Representations for Privacy-Aware Robot Perception
- The University of Texas at Austin（得克萨斯大学奥斯汀分校）
- Bogazici University（博加齐大学）
- Georgia Institute of Technology（佐治亚理工学院）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
RoboShape 是一种信息论引导的点云压缩工具，可在保留物体分类效用的同时降低敏感属性泄露，缩小嵌入体积，发布代码库供机器人社区使用。
AI中文摘要：
随着在人类环境中运行的智能体机器人通过扫描和共享三维表示（例如用于 fleet learning、基于云的规划或协作建图）的应用日益广泛，所收集的点云不仅会揭示场景中的物体，还会暴露敏感的空间上下文，例如房间功能或居住者从未同意披露的信息。传统的点云编码器对此没有原则性的控制：要么全部保留，要么全部丢弃。因此，我们引入了 RoboShape，这是一种遵循冻结 Sonata 编码器的信息论引导压缩头。我们使用互信息（MI）的 Donsker-Varadhan 公式来投影体素级嵌入，具体而言，我们最大化嵌入与物体级理解之间的互信息，同时最小化其与敏感属性之间的互信息。RoboShape 生成的嵌入体积缩小了 87.5%，同时保留了 98.7% 的物体分类效用，并且在三个真实世界室内 LiDAR 数据集上，敏感属性预测的坍塌程度达到了 39.3%。其隐私保护型嵌入更便于通过网络传输，也更便于为任何下游任务训练模型。我们发布了 RoboShape 代码库，为机器人社区提供了一种实用的、与编码器无关的工具，用于构建紧凑、隐私感知且可部署的感知流水线。
英文摘要：
With the increased adoption of robotic agents operating in human environments by scanning and sharing 3D representations (e.g., for fleet learning, cloud-based planning, or collaborative mapping), collected point clouds reveal not just the objects in a scene but also sensitive spatial context, such as room function or information that occupants never consented to disclose. Traditional point cloud encoders offer no principled control over this: either all is preserved, or none. Hence, we introduce RoboShape, an information theory guided compression head following the frozen {\tt Sonata} encoder. We project voxel-level embeddings using the Donsker-Varadhan formulation of mutual information (MI). Specifically, we maximize the MI between embeddings and object-level understanding while minimizing it for private attributes. RoboShape leads to 87.5\% smaller embeddings that retain 98.7\% of object classification utility while collapsing sensitive attribute predictions by 39.3\% across the three real-world indoor LiDAR datasets. Its privacy-preserving embeddings are cheaper to transmit over the network or to train a model for any downstream tasks. We release the RoboShape codebase to give the robotics community a practical, encoder-agnostic tool for building perception pipelines that are compact, privacy-aware, and deployment-ready.
