Category-wise Attack: Transferable Adversarial Examples for Anchor Free Object Detection

Abstract

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.

References in corpus (13)

• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
• Towards Deep Learning Models Resistant to Adversarial Attacks
• Robust Physical-World Attacks on Deep Learning Models
• DenseBox: Unifying Landmark Localization with End to End Object Detection
• A study of the effect of JPG compression on adversarial images
• Adversarial Transformation Networks: Learning to Generate Adversarial Examples
• Improving Adversarial Robustness via Promoting Ensemble Diversity
• Minimally distorted Adversarial Examples with a Fast Adaptive Boundary Attack
• Towards Interpretable Deep Neural Networks by Leveraging Adversarial Examples
• Robust Adversarial Perturbation on Deep Proposal-based Models
• Adversarial Images for Variational Autoencoders
• Transferable Adversarial Attacks for Image and Video Object Detection
• MeshAdv: Adversarial Meshes for Visual Recognition