Defending Against Physical Adversarial Patch Attacks on Infrared Human Detection
This addresses a safety-critical problem for infrared detection systems in real-world applications, though it appears incremental as it adapts existing defense concepts to a new domain.
The paper tackles the vulnerability of infrared human detection systems to physical adversarial patch attacks by proposing a defense strategy called patch-based occlusion-aware detection (POD), which robustly detects people and identifies adversarial patches while improving detection precision even in clean scenarios.
Infrared detection is an emerging technique for safety-critical tasks owing to its remarkable anti-interference capability. However, recent studies have revealed that it is vulnerable to physically-realizable adversarial patches, posing risks in its real-world applications. To address this problem, we are the first to investigate defense strategies against adversarial patch attacks on infrared detection, especially human detection. We propose a straightforward defense strategy, patch-based occlusion-aware detection (POD), which efficiently augments training samples with random patches and subsequently detects them. POD not only robustly detects people but also identifies adversarial patch locations. Surprisingly, while being extremely computationally efficient, POD easily generalizes to state-of-the-art adversarial patch attacks that are unseen during training. Furthermore, POD improves detection precision even in a clean (i.e., no-attack) situation due to the data augmentation effect. Our evaluation demonstrates that POD is robust to adversarial patches of various shapes and sizes. The effectiveness of our baseline approach is shown to be a viable defense mechanism for real-world infrared human detection systems, paving the way for exploring future research directions.