Efficient Task-Driven Video Data Privacy Protection for Smart Camera Surveillance System-Reference-Cited by-同舟云学术

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

Published:2023-10-02 Issue: Volume: Page:
ISSN:1550-4859
Container-title:ACM Transactions on Sensor Networks
language:en
Short-container-title:ACM Trans. Sen. Netw.

Author:

Wang Zhiqiang¹,Hou Jiahui¹,Wu Guangyu¹,Liu Suyuan¹,Luo Puhan¹,Li Xiangyang¹

Affiliation:

1. University of Science and Technology of China, China

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.

Publisher

Association for Computing Machinery (ACM)

Subject

Computer Networks and Communications

Link

https://dl.acm.org/doi/pdf/10.1145/3625825

Reference40 articles.

1. Aditya Chattopadhay , Anirban Sarkar , Prantik Howlader , and Vineeth N Balasubramanian . 2018. Grad-cam++ : Generalized gradient-based visual explanations for deep convolutional networks. In 2018 IEEE winter conference on applications of computer vision . IEEE , 839–847. Aditya Chattopadhay, Anirban Sarkar, Prantik Howlader, and Vineeth N Balasubramanian. 2018. Grad-cam++: Generalized gradient-based visual explanations for deep convolutional networks. In 2018 IEEE winter conference on applications of computer vision. IEEE, 839–847.

2. Sandeep Dsouza , Victor Bahl , Lixiang Ao , and Landon P Cox . 2020 . Amadeus: Scalable, Privacy-Preserving Live Video Analytics. arXiv preprint arXiv:2011.05163(2020). Sandeep Dsouza, Victor Bahl, Lixiang Ao, and Landon P Cox. 2020. Amadeus: Scalable, Privacy-Preserving Live Video Analytics. arXiv preprint arXiv:2011.05163(2020).

3. CenterNet: Keypoint Triplets for Object Detection

4. Minghong Fang , Xiaoyu Cao , Jinyuan Jia , and Neil Gong . 2020 . Local model poisoning attacks to {Byzantine-Robust} federated learning . In 29th USENIX Security Symposium (USENIX Security 20) . 1605–1622. Minghong Fang, Xiaoyu Cao, Jinyuan Jia, and Neil Gong. 2020. Local model poisoning attacks to {Byzantine-Robust} federated learning. In 29th USENIX Security Symposium (USENIX Security 20). 1605–1622.

5. Junyi Feng , Songyuan Li , Xi Li , et al . 2020 . Taplab : A fast framework for semantic video segmentation tapping into compressed-domain knowledge. IEEE Transactions on Pattern Analysis and Machine Intelligence ( 2020). Junyi Feng, Songyuan Li, Xi Li, et al. 2020. Taplab: A fast framework for semantic video segmentation tapping into compressed-domain knowledge. IEEE Transactions on Pattern Analysis and Machine Intelligence (2020).