期刊
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
卷 31, 期 8, 页码 3025-3036出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TCSVT.2020.3035722
关键词
Image restoration; Task analysis; Convolution; Kernel; Computational modeling; Recurrent neural networks; Video deblurring; recursive neural network; temporal consistency
资金
- National Natural Science Foundation of China [61922064]
- Zhejiang Provincial Natural Science Foundation [LR17F030001, LQ19F020005]
- Project of Science and Technology Plans of Wenzhou City [C20170008, G20150017, ZG2017016]
Video deblurring is a challenging task that requires accurate modeling of spatial and temporal characteristics. This article proposes a method to simulate temporal information and restore frame details, introducing a new loss function to ensure temporal consistency in generated frames.
Video deblurring is still a challenging low-level vision task since spatio-temporal characteristics across both the spatial and temporal domains are difficult to model. In this article, to model the temporal information, we develop a non-local block which estimates inter-frame similarity and inter-frame difference. Specially, for modeling the spatial characteristics and restoring sharp frame details, we propose a recursive block that iteratively refines feature maps generated at the last iteration. In addition, a novel temporal loss function is introduced to ensure the temporal consistency of generated frames. Experimental results on public datasets demonstrate that our method achieves state-of-the-art performance both quantitatively and qualitatively.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据