Unified Adversarial Patch for Cross-modal Attacks in the Physical World
This work addresses security vulnerabilities in multi-modal sensor systems, such as those used in surveillance or autonomous vehicles, by showing that cross-modal attacks are feasible, though it is incremental as it builds on existing adversarial patch methods.
The paper tackles the problem of physical adversarial attacks failing in multi-modal security systems by proposing a unified adversarial patch that simultaneously fools visible and infrared object detectors, achieving attack success rates of 73.33% and 69.17% against YOLOv3 and Faster RCNN, respectively, and demonstrating effectiveness in real-world settings.
Recently, physical adversarial attacks have been presented to evade DNNs-based object detectors. To ensure the security, many scenarios are simultaneously deployed with visible sensors and infrared sensors, leading to the failures of these single-modal physical attacks. To show the potential risks under such scenes, we propose a unified adversarial patch to perform cross-modal physical attacks, i.e., fooling visible and infrared object detectors at the same time via a single patch. Considering different imaging mechanisms of visible and infrared sensors, our work focuses on modeling the shapes of adversarial patches, which can be captured in different modalities when they change. To this end, we design a novel boundary-limited shape optimization to achieve the compact and smooth shapes, and thus they can be easily implemented in the physical world. In addition, to balance the fooling degree between visible detector and infrared detector during the optimization process, we propose a score-aware iterative evaluation, which can guide the adversarial patch to iteratively reduce the predicted scores of the multi-modal sensors. We finally test our method against the one-stage detector: YOLOv3 and the two-stage detector: Faster RCNN. Results show that our unified patch achieves an Attack Success Rate (ASR) of 73.33% and 69.17%, respectively. More importantly, we verify the effective attacks in the physical world when visible and infrared sensors shoot the objects under various settings like different angles, distances, postures, and scenes.