Efficient Task-Driven Video Data Privacy Protection for Smart Camera Surveillance System

Abstract

As one of the most commonly used AIoT sensors, smart cameras and their supporting services, namely cloud video surveillance (CVS) systems have brought great convenience to people’s lives. Recent CVS providers use different machine learning (ML) techniques to improve their services (regarded as tasks) based on the uploaded video. However, uploading data to the CVS providers may cause severe privacy issues. Existing works that remove privacy information could not achieve a high trade-off between data usability and privacy because the importance of information varies with the task. In addition, it is challenging to design a real-time privacy protection mechanism, especially in resource-constraint smart cameras. In this work, we design a task-driven and efficient video privacy protection mechanism for a better trade-off between privacy and data usability. We use Class Activation Mapping to protect privacy while preserving data usability. To improve the efficiency, we utilize the motion vector and residual matrix produced during video codec. Our work outperforms the ROI-based methods in data protection while preserving data usability. The attack accuracy drops 70%, while the task accuracy is comparable to those without protection (within ± 4%). The average protection frame rate of the High Definition video can exceed 16 fps+ even on a CPU.