Adversarial Machine Learning Attacks Against Video Anomaly Detection Systems

TL;DR

This paper demonstrates that Wi-Fi deauthentication attacks can generate adversarial video data, significantly impairing the reliability of video anomaly detection systems by causing false alarms and hiding anomalies.

Contribution

It introduces a novel cyber-attack method using Wi-Fi deauthentication to create adversarial effects on video quality, impacting anomaly detection models.

Findings

Wi-Fi deauthentication can cause video freezing and resolution loss.

Adversarial effects lead to increased false alarms in detection systems.

Attackers can hide physical anomalies from surveillance systems.

Abstract

Anomaly detection in videos is an important computer vision problem with various applications including automated video surveillance. Although adversarial attacks on image understanding models have been heavily investigated, there is not much work on adversarial machine learning targeting video understanding models and no previous work which focuses on video anomaly detection. To this end, we investigate an adversarial machine learning attack against video anomaly detection systems, that can be implemented via an easy-to-perform cyber-attack. Since surveillance cameras are usually connected to the server running the anomaly detection model through a wireless network, they are prone to cyber-attacks targeting the wireless connection. We demonstrate how Wi-Fi deauthentication attack, a notoriously easy-to-perform and effective denial-of-service (DoS) attack, can be utilized to generate adversarial data for video anomaly detection systems. Specifically, we apply several effects caused by the Wi-Fi deauthentication attack on video quality (e.g., slow down, freeze, fast forward, low resolution) to the popular benchmark datasets for video anomaly detection. Our experiments with several state-of-the-art anomaly detection models show that the attackers can significantly undermine the reliability of video anomaly detection systems by causing frequent false alarms and hiding physical anomalies from the surveillance system.