Single-image super-resolution with multilevel residual attention network

被引:9
作者
Qin, Ding [1 ]
Gu, Xiaodong [1 ]
机构
[1] Fudan Univ, Dept Elect Engn, Shanghai 200433, Peoples R China
基金
中国国家自然科学基金;
关键词
Image super-resolution; Deep convolutional neural network; Attention mechanism; Residual learning; NEURAL-NETWORK;
D O I
10.1007/s00521-020-04896-6
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Recently, a great variety of image super-resolution (SR) algorithms based on convolutional neural network (CNN) have been proposed and achieved significant improvement. But how to restore more high-frequency details such as edges and textures is still an unsolved issue. The low-frequency information is similar in a pair of low-resolution and high-resolution images. So the SR model is supposed to pay more attention to the high-frequency features to restore more realistic images. But most CNN-based methods don't consider the different types of features and think the features in different channels and regions contribute equally to the reconstruction performance, which limits the representation capacity of the model. In the meantime, most of these deep networks only simply stack blocks like residual block, which only capture the local features. In this paper, we propose a deep multilevel residual attention network (MRAN) for image SR to focus on the high-frequency features and improve the flow of information. Specially, we propose a channel-wise attention module and a spatial attention module to rescale the channel-wise and spatial weights adaptively, which makes our MRAN focus more on the high-frequency information. Meanwhile, to improve the flow of information and ease the training process, the multilevel residual learning is adopted. Extensive experimental results on five benchmark datasets demonstrate that our MRAN is superior to those state-of-the-art methods for both accuracy and visual comparisons.
引用
收藏
页码:15615 / 15628
页数:14
相关论文
共 36 条
  • [1] Contour Detection and Hierarchical Image Segmentation
    Arbelaez, Pablo
    Maire, Michael
    Fowlkes, Charless
    Malik, Jitendra
    [J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2011, 33 (05) : 898 - 916
  • [2] Low-Complexity Single-Image Super-Resolution based on Nonnegative Neighbor Embedding
    Bevilacqua, Marco
    Roumy, Aline
    Guillemot, Christine
    Morel, Marie-Line Alberi
    [J]. PROCEEDINGS OF THE BRITISH MACHINE VISION CONFERENCE 2012, 2012,
  • [3] A modular neural network for super-resolution of human faces
    Carcenac, Manuel
    [J]. APPLIED INTELLIGENCE, 2009, 30 (02) : 168 - 186
  • [4] Accelerating the Super-Resolution Convolutional Neural Network
    Dong, Chao
    Loy, Chen Change
    Tang, Xiaoou
    [J]. COMPUTER VISION - ECCV 2016, PT II, 2016, 9906 : 391 - 407
  • [5] Image Super-Resolution Using Deep Convolutional Networks
    Dong, Chao
    Loy, Chen Change
    He, Kaiming
    Tang, Xiaoou
    [J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2016, 38 (02) : 295 - 307
  • [6] Image Super-Resolution via Dual-State Recurrent Networks
    Han, Wei
    Chang, Shiyu
    Liu, Ding
    Yu, Mo
    Witbrock, Michael
    Huang, Thomas S.
    [J]. 2018 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2018, : 1654 - 1663
  • [7] Deep Back-Projection Networks For Super-Resolution
    Haris, Muhammad
    Shakhnarovich, Greg
    Ukita, Norimichi
    [J]. 2018 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2018, : 1664 - 1673
  • [8] He K., 2016, Deep residual learning for image recognition, DOI [10.1007/978-3-319-46493-0_38, DOI 10.1007/978-3-319-46493-0_38, DOI 10.1109/CVPR.2016.90]
  • [9] Hu J, 2018, PROC CVPR IEEE, P7132, DOI [10.1109/CVPR.2018.00745, 10.1109/TPAMI.2019.2913372]
  • [10] Huang G., 2017, Computer Vision and Pattern Recognition CVPR, P4700, DOI [DOI 10.1109/CVPR.2017.243, 10.1109/CVPR.2017.243]