Knowledge Commons of Institute of Automation,CAS
BiN-Flow: Bidirectional Normalizing Flow for Robust Image Dehazing | |
Wu, Yiqiang1,2; Tao, Dapeng1,3; Zhan, Yibing4; Zhang, Chenyang5 | |
发表期刊 | IEEE TRANSACTIONS ON IMAGE PROCESSING |
ISSN | 1057-7149 |
2022 | |
卷号 | 31页码:6635-6648 |
通讯作者 | Tao, Dapeng(dptao@ynu.edu.cn) |
摘要 | Image dehazing aims to remove haze in images to improve their image quality. However, most image dehazing methods heavily depend on strict prior knowledge and paired training strategy, which would hinder generalization and performance when dealing with unseen scenes. In this paper, to address the above problem, we propose Bidirectional Normalizing Flow (BiN-Flow), which exploits no prior knowledge and constructs a neural network through weakly-paired training with better generalization for image dehazing. Specifically, BiN-Flow designs 1) Feature Frequency Decoupling (FFD) for mining the various texture details through multi-scale residual blocks and 2) Bidirectional Propagation Flow (BPF) for exploiting the one-to-many relationships between hazy and haze-free images using a sequence of invertible Flow. In addition, BiN-Flow constructs a reference mechanism (RM) that uses a small number of paired hazy and haze-free images and a large number of haze-free reference images for weakly-paired training. Essentially, the mutual relationships between hazy and haze-free images could be effectively learned to further improve the generalization and performance for image dehazing. We conduct extensive experiments on five commonly-used datasets to validate the BiN-Flow. The experimental results that BiN-Flow outperforms all state-of-the-art competitors demonstrate the capability and generalization of our BiN-Flow. Besides, our BiN-Flow could produce diverse dehazing images for the same image by considering restoration diversity. |
关键词 | Training Image restoration Image color analysis Band-pass filters Convolutional neural networks Atmospheric modeling Noise measurement Image dehazing bidirectional normalizing flow weakly-paired training and restoration diversity |
DOI | 10.1109/TIP.2022.3214093 |
关键词[WOS] | RESTORATION ; VISIBILITY ; NETWORK ; WEATHER |
收录类别 | SCI |
语种 | 英语 |
资助项目 | Yunnan Provincial Major Science and Technology Special Plan Projects[202202AD080003] ; National Natural Science Foundation of China[62172354] ; National Natural Science Foundation of China[62002090] ; National Natural Science Foundation of China[62206293] ; Yunnan Natural Science Funds[202101AS070047] ; Yunnan Natural Science Funds[202205AG070003] ; Yunnan Natural Science Funds[2019FA-045] |
项目资助者 | Yunnan Provincial Major Science and Technology Special Plan Projects ; National Natural Science Foundation of China ; Yunnan Natural Science Funds |
WOS研究方向 | Computer Science ; Engineering |
WOS类目 | Computer Science, Artificial Intelligence ; Engineering, Electrical & Electronic |
WOS记录号 | WOS:000873807400005 |
出版者 | IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC |
引用统计 | |
文献类型 | 期刊论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/50515 |
专题 | 多模态人工智能系统全国重点实验室_互联网大数据与信息安全 |
通讯作者 | Tao, Dapeng |
作者单位 | 1.Yunnan Univ, Sch Informat Sci & Engn, FIST Lab, Kunming 650500, Yunnan, Peoples R China 2.Yunnan Key Lab Intelligent Syst & Comp, Kunming 650500, Yunnan, Peoples R China 3.Yunnan United Vis Technol Co Ltd, Kunming 650500, Yunnan, Peoples R China 4.JD Explore Acad, Beijing 100000, Peoples R China 5.Chinese Acad Sci, Inst Automat, State Key Lab Management & Control Complex Syst, Beijing 100190, Peoples R China |
推荐引用方式 GB/T 7714 | Wu, Yiqiang,Tao, Dapeng,Zhan, Yibing,et al. BiN-Flow: Bidirectional Normalizing Flow for Robust Image Dehazing[J]. IEEE TRANSACTIONS ON IMAGE PROCESSING,2022,31:6635-6648. |
APA | Wu, Yiqiang,Tao, Dapeng,Zhan, Yibing,&Zhang, Chenyang.(2022).BiN-Flow: Bidirectional Normalizing Flow for Robust Image Dehazing.IEEE TRANSACTIONS ON IMAGE PROCESSING,31,6635-6648. |
MLA | Wu, Yiqiang,et al."BiN-Flow: Bidirectional Normalizing Flow for Robust Image Dehazing".IEEE TRANSACTIONS ON IMAGE PROCESSING 31(2022):6635-6648. |
条目包含的文件 | 条目无相关文件。 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论