Summary of Itpatch: An Invisible and Triggered Physical Adversarial Patch Against Traffic Sign Recognition, by Shuai Yuan et al.
ITPatch: An Invisible and Triggered Physical Adversarial Patch against Traffic Sign Recognition
by Shuai Yuan, Hongwei Li, Xingshuo Han, Guowen Xu, Wenbo Jiang, Tao Ni, Qingchuan Zhao, Yuguang Fang
First submitted to arxiv on: 19 Sep 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This paper introduces an invisible and triggered physical adversarial patch (ITPatch) that uses fluorescent ink to advance the state-of-the-art in traffic sign recognition (TSR) system attacks. The ITPatch applies carefully designed fluorescent perturbations to a target sign, which can be later triggered using invisible ultraviolet light to cause misclassification of the TSR system and potentially result in traffic accidents. The paper evaluates the effectiveness of ITPatch, showing a success rate of 98.31% in low-light conditions and successfully bypassing five popular defenses with a success rate of 96.72%. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper creates an invisible patch that can trick self-driving cars to misread signs at night. It uses special ink that only shows up under certain lights, making it hard for the car’s computer to recognize the sign correctly. The authors tested this attack and found that it works most of the time, even when using special protection systems. |
