A Two-Stage Attentive Network for Single Image Super-Resolution

被引:43
|
作者
Zhang, Jiqing [1 ]
Long, Chengjiang [2 ]
Wang, Yuxin [1 ]
Piao, Haiyin [1 ]
Mei, Haiyang [3 ]
Yang, Xin [1 ]
Yin, Baocai [1 ]
机构
[1] Dalian Univ Technol, Dept Elect Informat & Elect Engn, Dalian 116024, Peoples R China
[2] JD Finance Amer Corp, San Francisco, CA 94043 USA
[3] Northwestern Polytech Univ, Sch Elect & Informat, Xian 710072, Peoples R China
基金
中国国家自然科学基金;
关键词
Feature extraction; Image reconstruction; Superresolution; Data mining; Ions; Convolution; Training; Single image super-resolution; deep learning; attention mechanism; multi-context block; two-stage; cross-dimension interaction; QUALITY ASSESSMENT; ACCURATE;
D O I
10.1109/TCSVT.2021.3071191
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
Recently, deep convolutional neural networks (CNNs) have been widely explored in single image super-resolution (SISR) and contribute remarkable progress. However, most of the existing CNNs-based SISR methods do not adequately explore contextual information in the feature extraction stage and pay little attention to the final high-resolution (HR) image reconstruction step, hence hindering the desired SR performance. To address the above two issues, in this paper, we propose a two-stage attentive network (TSAN) for accurate SISR in a coarse-to-fine manner. Specifically, we design a novel multi-context attentive block (MCAB) to make the network focus on more informative contextual features. Moreover, we present an essential refined attention block (RAB) which could explore useful cues in HR space for reconstructing fine-detailed HR image. Extensive evaluations on four benchmark datasets demonstrate the efficacy of our proposed TSAN in terms of quantitative metrics and visual effects. Code is available at https://github.com/Jee-King/TSAN.
引用
收藏
页码:1020 / 1033
页数:14
相关论文
共 50 条
  • [1] Two-stage Network For Single Image Super-Resolution
    Han, Yuzhuo
    Du, Xiaobiao
    Yang, Zhi
    [J]. 2021 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION WORKSHOPS, CVPRW 2021, 2021, : 880 - 887
  • [2] Two-Stage Convolutional Network for Image Super-Resolution
    Hui, Zheng
    Wang, Xiumei
    Gao, Xinbo
    [J]. 2018 24TH INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION (ICPR), 2018, : 2670 - 2675
  • [3] TFEN: two-stage feature enhancement network for single-image super-resolution
    Huang, Shuying
    Lai, Houzeng
    Yang, Yong
    Wan, Weiguo
    Li, Wei
    [J]. INTERNATIONAL JOURNAL OF MACHINE LEARNING AND CYBERNETICS, 2024, 15 (02) : 605 - 619
  • [4] TFEN: two-stage feature enhancement network for single-image super-resolution
    Shuying Huang
    Houzeng Lai
    Yong Yang
    Weiguo Wan
    Wei Li
    [J]. International Journal of Machine Learning and Cybernetics, 2024, 15 : 605 - 619
  • [5] Lightweight single image super-resolution with attentive residual refinement network
    Qin, Jinghui
    Zhang, Rumin
    [J]. NEUROCOMPUTING, 2022, 500 : 846 - 855
  • [6] Balanced Two-Stage Residual Networks for Image Super-Resolution
    Fan, Yuchen
    Shi, Honghui
    Yu, Jiahui
    Liu, Ding
    Han, Wei
    Yu, Haichao
    Wang, Zhangyang
    Wang, Xinchao
    Huang, Thomas S.
    [J]. 2017 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION WORKSHOPS (CVPRW), 2017, : 1157 - 1164
  • [7] An Improved Two-Stage Image Registration Algorithm for Super-Resolution
    Li, Xiangguo
    [J]. IEEJ TRANSACTIONS ON ELECTRICAL AND ELECTRONIC ENGINEERING, 2014, 9 (04) : 415 - 420
  • [8] A Two-Stage Three-Dimensional Attention Network for Lightweight Image Super-Resolution
    Chen, Lei
    Yang, Yanjie
    Zhuang, Xu
    Wang, Jason
    Mao, Qin
    Yue, Hong
    Wei, Xuekai
    Cheng, Fei
    Zong, Xuemei
    Zhou, Mingliang
    [J]. INTERNATIONAL JOURNAL OF PATTERN RECOGNITION AND ARTIFICIAL INTELLIGENCE, 2023, 37 (13)
  • [9] Image Super-Resolution Reconstruction Based on Two-Stage Dictionary Learning
    Shang, Li
    Sun, Zhan-li
    [J]. INTELLIGENT COMPUTING METHODOLOGIES, 2014, 8589 : 277 - 284
  • [10] Super-resolution reconstruction based on two-stage residual neural network
    Dong, Lin
    Inoue, Kohei
    [J]. MACHINE LEARNING WITH APPLICATIONS, 2021, 6