APRICOT: A Dataset of Physical Adversarial Attacks on Object Detection
This work addresses the need for reproducible research on physical adversarial attacks for computer vision researchers, though it is incremental as it builds on existing datasets and methods.
The authors tackled the problem of evaluating physical adversarial attacks on object detection systems by creating APRICOT, a dataset of over 1,000 annotated photos of adversarial patches in real-world settings, and found that while maintaining robustness is challenging, patches can be effectively flagged using both supervised and unsupervised defense methods.
Physical adversarial attacks threaten to fool object detection systems, but reproducible research on the real-world effectiveness of physical patches and how to defend against them requires a publicly available benchmark dataset. We present APRICOT, a collection of over 1,000 annotated photographs of printed adversarial patches in public locations. The patches target several object categories for three COCO-trained detection models, and the photos represent natural variation in position, distance, lighting conditions, and viewing angle. Our analysis suggests that maintaining adversarial robustness in uncontrolled settings is highly challenging, but it is still possible to produce targeted detections under white-box and sometimes black-box settings. We establish baselines for defending against adversarial patches through several methods, including a detector supervised with synthetic data and unsupervised methods such as kernel density estimation, Bayesian uncertainty, and reconstruction error. Our results suggest that adversarial patches can be effectively flagged, both in a high-knowledge, attack-specific scenario, and in an unsupervised setting where patches are detected as anomalies in natural images. This dataset and the described experiments provide a benchmark for future research on the effectiveness of and defenses against physical adversarial objects in the wild.