W-Net: Structure and Texture Interaction for Image Inpainting
文献类型:期刊论文
作者 | Zhang, Ruisong1,3; Quan, Weize1,3; Zhang, Yong2; Wang, Jue2; Yan, Dong-Ming1,3 |
刊名 | IEEE Transactions on Multimedia |
出版日期 | 2022-11 |
页码 | 1-12 |
英文摘要 | Recent literature has developed two advanced tools for image inpainting: appearance propagation and attention matching. However, given the ineffective feature reorganization and vulnerable attention maps, existing works yield suboptimal results with distorted structures and inconsistent contents. Furthermore, we observe that deep sampling layers (DSL) and shallow skip connections (SSC) in U-Net separately promote image structure inference and texture synthesis. To address the above two issues, we devise a W-shaped network (W-Net), which consists of two key components: a texture spatial attention (TSA) module in SSC and a structure channel excitation (SCE) module in DSL. W-Net is a two-stage network, with coarse and refined structures derived at each stage. Meanwhile, the TSA module fills incomplete textures with reliable attention scores under the guidance of coarse structures, which effectively diminishes inconsistency from appearance to semantics. The SCE module rectifies structures according to the difference between coarse structures and refined structures enhanced by texture features. Then the module motivates them to produce more reasonable shapes. Complete textures and refined structures constitute desired inpainted images, as the output of W-Net. Experiments on multiple datasets demonstrate the superior performance of W-Net. The source code is available at https://github.com/Evergrow/W-Net |
源URL | [http://ir.ia.ac.cn/handle/173211/51500] |
专题 | 多模态人工智能系统全国重点实验室 |
通讯作者 | Yan, Dong-Ming |
作者单位 | 1.School of Artificial Intelligence, the University of Chinese Academy of Sciences 2.Tencent AI Lab, ShenZhen 3.NLPR, Institute of Automation, Chinese Academy of Science |
推荐引用方式 GB/T 7714 | Zhang, Ruisong,Quan, Weize,Zhang, Yong,et al. W-Net: Structure and Texture Interaction for Image Inpainting[J]. IEEE Transactions on Multimedia,2022:1-12. |
APA | Zhang, Ruisong,Quan, Weize,Zhang, Yong,Wang, Jue,&Yan, Dong-Ming.(2022).W-Net: Structure and Texture Interaction for Image Inpainting.IEEE Transactions on Multimedia,1-12. |
MLA | Zhang, Ruisong,et al."W-Net: Structure and Texture Interaction for Image Inpainting".IEEE Transactions on Multimedia (2022):1-12. |
入库方式: OAI收割
来源:自动化研究所
其他版本
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。