VALSE 论文速览 第105期：Generative SEIDNet for Image Rain Removal

论文题目：Generative Status Estimation and Information Decoupling for Image Rain Removal

作者列表：林迪 (天津大学)， 王鑫 (香港理工大学)，申佳 (天津大学)，张仁杰 (香港理工大学)， 刘若楠 (天津大学)，王妙辉 (深圳大学)，谢伍媛 (深圳大学)，郭青 (新加坡科技研究局)，李平 (香港理工大学)

B站观看网址：

https://www.bilibili.com/video/BV13Y411671b/

论文摘要：

Image rain removal requires the accurate separation between the pixels of the rain streaks and object textures. But the confusing appearances of rains and objects lead to the misunderstanding of pixels, thus remaining the rain streaks or missing the object details in the result. This talk will present the generative Status Estimation and Information Decoupling for rain removal. We will also discuss how the generative networks may benefit the various image restoration tasks.

论文信息：

[1] Di Lin, Xin Wang, Jia Shen, Renjie Zhang, Ruonan Liu, Miaohui Wang, Wuyuan Xie, Qing Guo, and Ping Li, “Generative Status Estimation and Information Decoupling for Image Rain Removal,” Neural Information Processing Systems (NeurIPS), 2022.

论文链接：

[https://openreview.net/pdf?id=C2o5DeL_8L1]

视频讲者简介：

Xin Wang is a Ph.D. Candidate at the Department of Computing of The Hong Kong Polytechnic University, under the supervision of Prof. Ping Li. He received the bachelor's degree in Computer Science and Technology from Dalian University of Technology in 2017. His research interests contain deep learning and image synthesis.

特别鸣谢本次论文速览主要组织者：

月度轮值AC：林迪 (天津大学)、胡鹏 (四川大学)