PACAS: A Privacy-Aware Smart Camera System

• Published in: 2024 IEEE Cloud Summit pp. 170 - 177
• Main Authors: Yu, Keyang, Chen, Dong
• Format: Conference Proceeding
• Language: English
• Published: IEEE 27-06-2024
• Subjects: Cloud computing; Internet of Things; Privacy; Prototypes; Recording; Smart cameras; Smart homes; Streaming media; Tiny ML; User Privacy
• DOI: 10.1109/Cloud-Summit61220.2024.00035

Smart cameras have been increasingly deployed in smart homes for remote monitoring and enhancing home security. However, extensive recent research has uncovered potential user privacy threats associated with popular commercial camera systems. Some manufacture design of these commercial camera systems usually requires smart camera users to relinquish their control of camera recorded data. For instance, these cameras often upload camera recordings to their cloud servers to enable advanced data analysis for camera app services. To facilitate enhanced camera services, the data may be further shared with on-path vendors, third parties of manufacturers, and cloud providers, potentially allowing them to access video footage or image captures without users' awareness or meaningful consent. To address this problem, we design a new smart camera system-PACAS that enables people to regain the control of their data while still retaining access to regular camera services. We evaluate PACAS using multiple camera video footage traces and on multiple real-world camera prototypes. We show that PACAS can achieve the performance of 0.5 second living streaming latency at a frame rate of 30 fps and a resolution of 240x320 on average. PACAS offers compelling evidence that smart camera systems can deliver on-device modern capabilities comparable to those provided by current commercial systems, all while upholding user privacy.