Fooling Object Detectors: Adversarial Attacks by Half-Neighbor Masks
arXiv:2101.00989
Abstract
Although there are a great number of adversarial attacks on deep learning based classifiers, how to attack object detection systems has been rarely studied. In this paper, we propose a Half-Neighbor Masked Projected Gradient Descent (HNM-PGD) based attack, which can generate strong perturbation to fool different kinds of detectors under strict constraints. We also applied the proposed HNM-PGD attack in the CIKM 2020 AnalytiCup Competition, which was ranked within the top 1% on the leaderboard. We release the code at https://github.com/YanghaoZYH/HNM-PGD.
arXiv commentTo appear in the Proceedings of the CIKM 2020 Workshops published by CEUR-WS
References in corpus (7)
- YOLOv4: Optimal Speed and Accuracy of Object Detection
- SmoothGrad: removing noise by adding noise
- A Game-Based Approximate Verification of Deep Neural Networks with Provable Guarantees
- Adversarial Examples that Fool Detectors
- Reachability Analysis of Deep Neural Networks with Provable Guarantees
- Cooling-Shrinking Attack: Blinding the Tracker with Imperceptible Noises
- Generalizing Universal Adversarial Attacks Beyond Additive Perturbations