Summary of Pb-uap: Hybrid Universal Adversarial Attack For Image Segmentation, by Yufei Song et al.
PB-UAP: Hybrid Universal Adversarial Attack For Image Segmentation
by Yufei Song, Ziqi Zhou, Minghui Li, Xianlong Wang, Hangtao Zhang, Menghao Deng, Wei Wan, Shengshan Hu, Leo Yu Zhang
First submitted to arxiv on: 21 Dec 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 The proposed paper introduces a novel universal adversarial attack method for segmentation models, focusing on deep neural networks. The approach combines dual feature separation and low-frequency scattering modules to generate effective attacks in pixel and frequency spaces. Experimental results show high success rates, surpassing state-of-the-art methods, and demonstrate strong transferability across different models. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper is about making computer vision models less reliable by creating fake images that can trick them. Currently, most research focuses on image classification, but this study looks at segmentation models, which are used for tasks like road mapping or medical imaging. The method uses two techniques to create these “adversarial” images and shows it’s effective and can work with different models. |
Keywords
» Artificial intelligence » Image classification » Transferability
