Category-wise Attack: Transferable Adversarial Examples for Anchor Free Object Detection
This addresses a security problem for users of object detection systems, particularly in safety-critical applications, and is incremental as it builds on existing adversarial attack research.
The paper tackles the vulnerability of anchor-free object detection models to adversarial attacks by introducing a category-wise attack method that leverages high-level semantic information, achieving transferability to other detectors including anchor-based ones like Faster R-CNN.
Deep neural networks have been demonstrated to be vulnerable to adversarial attacks: subtle perturbations can completely change the classification results. Their vulnerability has led to a surge of research in this direction. However, most works dedicated to attacking anchor-based object detection models. In this work, we aim to present an effective and efficient algorithm to generate adversarial examples to attack anchor-free object models based on two approaches. First, we conduct category-wise instead of instance-wise attacks on the object detectors. Second, we leverage the high-level semantic information to generate the adversarial examples. Surprisingly, the generated adversarial examples it not only able to effectively attack the targeted anchor-free object detector but also to be transferred to attack other object detectors, even anchor-based detectors such as Faster R-CNN.