Defending Person Detection Against Adversarial Patch Attack by using Universal Defensive Frame
This addresses a critical security problem in safety-critical applications like autonomous driving and security systems, but it is an incremental improvement as it builds on existing defense strategies for adversarial attacks.
The paper tackles the vulnerability of person detection networks to adversarial patch attacks by proposing a universal defensive frame optimized through competitive learning, which effectively defends against attacks while maintaining detection performance on clean images.
Person detection has attracted great attention in the computer vision area and is an imperative element in human-centric computer vision. Although the predictive performances of person detection networks have been improved dramatically, they are vulnerable to adversarial patch attacks. Changing the pixels in a restricted region can easily fool the person detection network in safety-critical applications such as autonomous driving and security systems. Despite the necessity of countering adversarial patch attacks, very few efforts have been dedicated to defending person detection against adversarial patch attack. In this paper, we propose a novel defense strategy that defends against an adversarial patch attack by optimizing a defensive frame for person detection. The defensive frame alleviates the effect of the adversarial patch while maintaining person detection performance with clean person. The proposed defensive frame in the person detection is generated with a competitive learning algorithm which makes an iterative competition between detection threatening module and detection shielding module in person detection. Comprehensive experimental results demonstrate that the proposed method effectively defends person detection against adversarial patch attacks.