Fast Hierarchical Deep Unfolding Network for Image Compressed Sensing

被引:6
|
作者
Cui, Wenxue [1 ]
Liu, Shaohui [1 ,2 ]
Zhao, Debin [1 ,2 ]
机构
[1] Harbin Inst Technol, Harbin, Peoples R China
[2] Peng Cheng Lab, Shenzhen, Peoples R China
基金
中国国家自然科学基金;
关键词
Image compressed sensing; deep unfolding network (DUN); Fast Iterative Shrinkage-Thresholding Algorithm (FISTA); hierarchical architecture; hyperparametric generation; RECONSTRUCTION; ALGORITHMS;
D O I
10.1145/3503161.3548389
中图分类号
TP39 [计算机的应用];
学科分类号
081203 ; 0835 ;
摘要
By integrating certain optimization solvers with deep neural network, deep unfolding network (DUN) has attracted much attention in recent years for image compressed sensing (CS). However, there still exist several issues in existing DUNs: 1) For each iteration, a simple stacked convolutional network is usually adopted, which apparently limits the expressiveness of these models. 2) Once the training is completed, most hyperparameters of existing DUNs are fixed for any input content, which significantly weakens their adaptability. In this paper, by unfolding the Fast Iterative Shrinkage-Thresholding Algorithm (FISTA), a novel fast hierarchical DUN, dubbed FHDUN, is proposed for image compressed sensing, in which a well-designed hierarchical unfolding architecture is developed to cooperatively explore richer contextual prior information in multi-scale spaces. To further enhance the adaptability, series of hyperparametric generation networks are developed in our framework to dynamically produce the corresponding optimal hyperparameters according to the input content. Furthermore, due to the accelerated policy in FISTA, the newly embedded acceleration module makes the proposed FHDUN save more than 50% of the iterative loops against recent DUNs. Extensive CS experiments manifest that the proposed FHDUN outperforms existing state-of-the-art CS methods, while maintaining fewer iterations.
引用
收藏
页码:2739 / 2748
页数:10
相关论文
共 50 条
  • [1] Gates-Controlled Deep Unfolding Network for Image Compressed Sensing
    Li, Tiancheng
    Yan, Qiurong
    Zou, Quan
    Dai, Qianling
    [J]. IEEE TRANSACTIONS ON COMPUTATIONAL IMAGING, 2024, 10 : 103 - 114
  • [2] Gates-Controlled Deep Unfolding Network for Image Compressed Sensing
    Li, Tiancheng
    Yan, Qiurong
    Zou, Quan
    Dai, Qianling
    [J]. IEEE Transactions on Computational Imaging, 2024, 10 : 103 - 114
  • [3] Fast bilateral complementary network for deep learning compressed sensing image reconstruction
    Guo Yuan
    Jiang Jinlin
    Chen Wei
    [J]. IET IMAGE PROCESSING, 2022, 16 (13) : 3485 - 3498
  • [4] WDU-Net: Wavelet-Guided Deep Unfolding Network for Image Compressed Sensing Reconstruction
    Wang, Xinlu
    Zhao, Lijun
    Zhang, Jinjing
    Zhang, Yufeng
    Wang, Anhong
    [J]. PATTERN RECOGNITION AND COMPUTER VISION, PRCV 2023, PT VI, 2024, 14430 : 79 - 91
  • [5] ADMM-DAD NET: A DEEP UNFOLDING NETWORK FOR ANALYSIS COMPRESSED SENSING
    Kouni, Vasiliki
    Paraskevopoulos, Georgios
    Rauhut, Holger
    Alexandropoulos, George C.
    [J]. 2022 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP), 2022, : 1506 - 1510
  • [6] A Deep Network Based on Wavelet Transform for Image Compressed Sensing
    Zhu Yin
    Zhongcheng Wu
    Jun Zhang
    [J]. Circuits, Systems, and Signal Processing, 2022, 41 : 6031 - 6050
  • [7] Adaptive deep learning network for image reconstruction of compressed sensing
    Nan, Ruili
    Sun, Guiling
    Zheng, Bowen
    Wang, Lin
    [J]. SIGNAL IMAGE AND VIDEO PROCESSING, 2024, 18 (02) : 1463 - 1475
  • [8] DMFNet: deep matrix factorization network for image compressed sensing
    Wang, Hengyou
    Li, Haocheng
    Jiang, Xiang
    [J]. MULTIMEDIA SYSTEMS, 2024, 30 (04)
  • [9] A Deep Network Based on Wavelet Transform for Image Compressed Sensing
    Yin, Zhu
    Wu, Zhongcheng
    Zhang, Jun
    [J]. CIRCUITS SYSTEMS AND SIGNAL PROCESSING, 2022, 41 (11) : 6031 - 6050
  • [10] Adaptive deep learning network for image reconstruction of compressed sensing
    Ruili Nan
    Guiling Sun
    Bowen Zheng
    Lin Wang
    [J]. Signal, Image and Video Processing, 2024, 18 : 1463 - 1475