Privacy-Aware Sharing of Raw Spatial Sensor Data for Cooperative Perception

2025-12-18

AI Summary

To address novel privacy risks arising from raw-sensor-data sharing in vehicle-infrastructure cooperative perception, this paper proposes SHARP—a framework that systematically models and empirically validates privacy threats for such spatial data for the first time. Unlike prior approaches relying on data anonymization or model distillation, SHARP introduces a privacy metric model, lightweight spatial-data perturbation, and cross-vehicle privacy–utility co-optimization—enabling fine-grained, controllable privacy protection while preserving raw-data usability. Experiments on real-world LiDAR and camera datasets demonstrate that SHARP reduces privacy leakage by 62% compared to baseline methods, with a marginal degradation of less than 1.5% in collaborative perception mean Average Precision (mAP). Thus, SHARP achieves a significant balance between rigorous privacy guarantees and high perceptual performance.

Technology Category

• Machine Learning: Privacy
• Computer Vision: Bias, Fairness & Privacy
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Security and Privacy: Data transparency and provenance
• User Modeling, Personalization and Recommendation: User privacy protection in personalized systems
• Responsible Web: Data and user privacy-enhancing technologies for the Web

Abstract

Cooperative perception between vehicles is poised to offer robust and reliable scene understanding. Recently, we are witnessing experimental systems research building testbeds that share raw spatial sensor data for cooperative perception. While there has been a marked improvement in accuracies and is the natural way forward, we take a moment to consider the problems with such an approach for eventual adoption by automakers. In this paper, we first argue that new forms of privacy concerns arise and discourage stakeholders to share raw sensor data. Next, we present SHARP, a research framework to minimize privacy leakage and drive stakeholders towards the ambitious goal of raw data based cooperative perception. Finally, we discuss open questions for networked systems, mobile computing, perception researchers, industry and government in realizing our proposed framework.

Problem

• Addresses privacy concerns in raw spatial sensor data sharing
• Proposes SHARP framework to minimize privacy leakage risks
• Encourages adoption of raw data for cooperative perception systems

Innovation

• SHARP framework minimizes privacy leakage
• Addresses privacy concerns in raw sensor data sharing
• Enables cooperative perception with raw spatial data