VALSE 论文速览 第203期：针对图像去雾模型的攻击研究

论文题目：

Fooling the Image Dehazing Models by First Order Gradient

作者列表：

桂杰 (东南大学)，丛晓峰 (东南大学)，彭程威 (腾讯深圳公司)，唐遠炎 (澳门大学)，郭天佑 (香港科技大学)

B站观看网址：

https://www.bilibili.com/video/BV1oXxNeAEqb/?spm_id_from=333.999.0.0&vd_source=2044ac986b5caa4c8b5f00b525441df3

复制链接到浏览器打开或点击阅读原文即可跳转至观看页面。

论文摘要：

The research on the single image dehazing task has been widely explored. However, as far as we know, no comprehensive study has been conducted on the robustness of the well-trained dehazing models. In this paper, we focus on designing a group of attack methods based on first-order gradient to verify the robustness of the existing dehazing algorithms. Further, the defense strategy based on adversarial training is adopted for reducing the negative effects caused by malicious attacks. In summary, this paper defines a new challenging problem for the image dehazing area, which can be called as adversarial attack on dehazing networks (AADN). 

参考文献：

[1] Jie Gui, Xiaofeng Cong, Chengwei Peng, Yuan Yan Tang, James Tin-Yau Kwok. Fooling the Image Dehazing Models by First Order Gradient. IEEE Transactions on Circuits and Systems for Video Technology, 2024.

论文链接：

[https://arxiv.org/abs/2303.17255]

代码链接：

[https://github.com/Xiaofeng-life/AADN_Dehazing]

视频讲者简介：

Xiaofeng Cong is pursuing his Ph.D. degree in the School of Cyber Science and Engineering, Southeast University, Nanjing, China. He has published papers in CVPR, AAAI, IJCAI, ACM MM, ACM CSUR, TMM, TCSVT and TGRS, etc. His research interests include underwater image enhancement and image dehazing.

个人主页：

https://xiaofeng-life.github.io/index.html

特别鸣谢本次论文速览主要组织者：

月度轮值AC：武宇 (武汉大学)