arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2008.00455cs.CV
收藏

递归结构-细节网络视频超分辨率

Video Super-Resolution with Recurrent Structure-Detail Network

Tsinghua University(清华大学);The University of Sydney(悉尼大学)

Takashi Isobe, Xu Jia, Shuhang Gu, Songjiang Li, Shengjin Wang, Qi Tian

摘要:

大多数视频超分辨率方法在时间滑动窗口中的相邻帧的帮助下对单个参考帧进行超分辨率。与基于递归的方法相比,它们的效率较低。在这项工作中,我们提出了一种新的经常性的视频超分辨率方法,这是有效的和高效的利用以前的帧超分辨率的当前帧。它将输入分为结构和细节分量,这些分量被馈送到由几个建议的双流结构细节块组成的循环单元。此外,隐藏状态自适应模块,允许当前帧选择性地使用隐藏状态的信息被引入,以提高其鲁棒性的外观变化和错误积累。广泛的消融研究验证了所提出的模块的有效性。在多个基准数据集上的实验表明,与现有的视频超分辨率方法相比,该方法具有更好的性能。

英文摘要:

Most video super-resolution methods super-resolve a single reference frame with the help of neighboring frames in a temporal sliding window. They are less efficient compared to the recurrent-based methods. In this work, we propose a novel recurrent video super-resolution method which is both effective and efficient in exploiting previous frames to super-resolve the current frame. It divides the input into structure and detail components which are fed to a recurrent unit composed of several proposed two-stream structure-detail blocks. In addition, a hidden state adaptation module that allows the current frame to selectively use information from hidden state is introduced to enhance its robustness to appearance change and error accumulation. Extensive ablation study validate the effectiveness of the proposed modules. Experiments on several benchmark datasets demonstrate the superior performance of the proposed method compared to state-of-the-art methods on video super-resolution.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑