基于高分辨率深度与超低分辨率RGB的隐私保护语义分割
Privacy-Preserving Semantic Segmentation from High-Resolution Depth and Ultra-Low-Resolution RGB
浏览论文内容
中文总结 AI 辅助
针对移动机器人视觉隐私问题，提出结合高分辨率深度与超低分辨率RGB的非对称感知设置，通过联合2D框架和2D到3D流水线实现隐私保护下的最佳2D/3D分割及零样本迁移，并验证了隐私可恢复性降低与导航实用性。
中文摘要 AI 辅助
随着移动机器人日益融入日常环境，由机载摄像头引发的隐私风险已成为日益关注的问题。超低分辨率（ULR）RGB可以在源头减轻视觉隐私暴露，但仅凭超低分辨率外观会大幅限制语义和空间理解。因此，我们引入了一种隐私保护的非对称感知设置，将高分辨率（HR）深度与超低分辨率RGB相结合，在保留密集几何信息的同时限制细粒度视觉信息。为了解决高分辨率深度与超低分辨率RGB之间的严重信息不平衡问题，我们提出了一个联合2D框架，利用高分辨率几何信息引导面向语义的RGB重建和RGB-D分割。尽管帧级预测可靠，但在非对称的HR深度-ULR RGB设置下，一致的场景级理解仍然具有挑战性。因此，我们开发了一个端到端的2D到3D流水线，整合2D语义特征以进行3D分割。在ScanNet上的实验表明，我们的方法在隐私保护方法中取得了最佳的2D和3D分割性能，并实现了对SUN RGB-D和SceneNN的最强零样本迁移。隐私可恢复性分析表明，我们提出的HR深度-ULR RGB输入降低了敏感数据的可恢复性，真实机器人实验证明了所得3D语义在目标导向导航中的实用性。
英文摘要
As mobile robots become increasingly integrated into everyday environments, privacy risks arising from onboard cameras have become a growing concern. Ultra-low-resolution (ULR) RGB can mitigate visual privacy exposure at the source, but ULR appearance alone substantially limits semantic and spatial understanding. We therefore introduce a privacy-preserving asymmetric sensing setting that combines high-resolution (HR) depth with ULR RGB, preserving dense geometry while restricting fine-grained visual information. To address the severe information imbalance between HR depth and ULR RGB, we propose a joint 2D framework using HR geometry to guide semantic-oriented RGB reconstruction and RGB-D segmentation. Despite reliable frame-level predictions, consistent scene-level understanding remains challenging under the asymmetric HR depth--ULR RGB setting. We therefore develop an end-to-end 2D-to-3D pipeline that consolidates 2D semantic features for 3D segmentation. Experiments on ScanNet show that our method achieves the best 2D and 3D segmentation performance among privacy-preserving approaches and delivers the strongest zero-shot transfer to SUN RGB-D and SceneNN. Privacy recoverability analysis shows that our proposed HR depth--ULR RGB input reduces the recoverability of sensitive data, and real-robot experiments demonstrate the utility of the resulting 3D semantics for object-goal navigation.
发表机构
- University of Bonn（波恩大学）
- Lamarr Institute for Machine Learning and Artificial Intelligence（拉马尔机器学习和人工智能研究所）
- Center for Robotics（机器人中心）
- Bonn-Rhein-Sieg University of Applied Sciences（波恩-莱茵-锡格应用科学大学）
- Fraunhofer Institute for Intelligent Analysis and Information Systems（弗劳恩霍夫智能分析和信息系统研究所）
机构由 AI 辅助整理，请以论文原文为准。
