文献类型：期刊文献

英文题名：Structure-Aware Motion Deblurring Using Multi-Adversarial Optimized CycleGAN

作者：Wen, Yang[1];Chen, Jie[2];Sheng, Bin[1];Chen, Zhihua[3];Li, Ping[4];Tan, Ping[5];Lee, Tong-Yee[6]

机构：[1]Shanghai Jiao Tong Univ, Dept Comp Sci & Engn, Shanghai 200240, Peoples R China;[2]Samsung Elect China Res & Dev Ctr, Nanjing 210012, Peoples R China;[3]East China Univ Sci & Technol, Dept Comp Sci & Engn, Shanghai 200237, Peoples R China;[4]Hong Kong Polytech Univ, Dept Comp, Hong Kong, Peoples R China;[5]Simon Fraser Univ, Sch Comp Sci, Burnaby, BC V5A 1S6, Canada;[6]Natl Cheng Kung Univ, Dept Comp Sci & Informat Engn, Tainan 70101, Taiwan

年份：2021

卷号：30

起止页码：6142

外文期刊名：IEEE TRANSACTIONS ON IMAGE PROCESSING

收录：;EI(收录号：20213210755093);WOS:【SCI-EXPANDED(收录号:WOS:000671507400002)】；

基金：This work was supported in part by the National Natural Science Foundation of China under Grant 61872241 and Grant 61572316; in part by The Hong Kong Polytechnic University under Grant P0030419, Grant P0030929, and Grant P0035358; and in part by the Ministry of Science and Technology, Taiwan, under Grant 108-2221-E-006-038-MY3.

语种：英文

外文关键词：Image edge detection; Kernel; Image restoration; Estimation; Training data; Generative adversarial networks; Computer architecture; Unsupervised image deblurring; multi-adversarial; structure-aware; edge refinement

摘要：Recently, Convolutional Neural Networks (CNNs) have achieved great improvements in blind image motion deblurring. However, most existing image deblurring methods require a large amount of paired training data and fail to maintain satisfactory structural information, which greatly limits their application scope. In this paper, we present an unsupervised image deblurring method based on a multi-adversarial optimized cycle-consistent generative adversarial network (CycleGAN). Although original CycleGAN can handle unpaired training data well, the generated high-resolution images are probable to lose content and structure information. To solve this problem, we utilize a multi-adversarial mechanism based on CycleGAN for blind motion deblurring to generate high-resolution images iteratively. In this multi-adversarial manner, the hidden layers of the generator are gradually supervised, and the implicit refinement is carried out to generate high-resolution images continuously. Meanwhile, we also introduce the structure-aware mechanism to enhance the structure and detail retention ability of the multi-adversarial network for deblurring by taking the edge map as guidance information and adding multi-scale edge constraint functions. Our approach not only avoids the strict need for paired training data and the errors caused by blur kernel estimation, but also maintains the structural information better with multi-adversarial learning and structure-aware mechanism. Comprehensive experiments on several benchmarks have shown that our approach prevails the state-of-the-art methods for blind image motion deblurring.