Seeing isn't Believing: Practical Adversarial Attack Against Object Detectors
This work addresses security vulnerabilities in object detection systems, which is critical for applications like autonomous vehicles and surveillance, but it is incremental as it builds on existing adversarial attack methods.
The paper tackled the problem of creating practical adversarial examples (AEs) to fool real-world object detectors, achieving success rates up to 92.4% against state-of-the-art models like YOLO V3 and faster-RCNN across varying distances and angles.
In this paper, we presented systematic solutions to build robust and practical AEs against real world object detectors. Particularly, for Hiding Attack (HA), we proposed the feature-interference reinforcement (FIR) method and the enhanced realistic constraints generation (ERG) to enhance robustness, and for Appearing Attack (AA), we proposed the nested-AE, which combines two AEs together to attack object detectors in both long and short distance. We also designed diverse styles of AEs to make AA more surreptitious. Evaluation results show that our AEs can attack the state-of-the-art real-time object detectors (i.e., YOLO V3 and faster-RCNN) at the success rate up to 92.4% with varying distance from 1m to 25m and angles from -60° to 60°. Our AEs are also demonstrated to be highly transferable, capable of attacking another three state-of-the-art black-box models with high success rate.