arrow
返回

MTC-CSNet: Marrying Transformer and Convolution for Image Compressed Sensing

delete2024-09-01
delete0
PRE
AI
M
Minghe Shen
H
Hongping Gan *
C
Chunyan Ma
C
Chao Ning
H
Hongqi Li
刘锋 封面图
刘锋 (Feng Liu)
DOI:10.1109/TCYB.2024.3363748delete
delete原文链接
delete原文求助
delete分享
delete收藏
摘要

摘要

En 中文
Image compressed sensing (ICS) has been extensively applied in various imaging domains due to its capability to sample and reconstruct images at subNyquist sampling rates. The current predominant approaches in ICS, specifically pure convolutional networks (ConvNets)-based ICS methods, have demonstrated their effectiveness in capturing local features for image recovery. Simultaneously, the Transformer architecture has gained significant attention due to its capability to model global correlations among image features. Motivated by these insights, we propose a novel hybrid network for ICS, named MTC-CSNet, which effectively combines the strengths of both ConvNets and Transformer architectures in capturing local and global image features to achieve high-quality image recovery. Particularly, MTC-CSNet is a dual-path framework that consists of a ConvNets-based recovery branch and a Transformer-based recovery branch. Along the ConvNets-based recovery branch, we design a lightweight scheme to capture the local features in natural images. Meanwhile, we implement a Transformer-based recovery branch to iteratively model the global dependencies among image patches. Ultimately, the ConvNets-based and Transformer-based recovery branches collaborate through a bridging unit, facilitating the adaptive transmission and fusion of informative features for image reconstruction. Extensive experimental results demonstrate that our proposed MTC-CSNet surpasses the state-of-the-art methods on various public datasets. The code and models are publicly available at MTC-CSNet.
Keyword:
Compressed sensing
convolutional networks (ConvNets)
deep learning (DL)
image recovery
transformer

期刊

IEEE Transactions on Cybernetics 封面图
IEEE Transactions on Cybernetics
IF:
10.5
论文数:
1.1W
被引数:
5.0W

机构

N
Northwestern Polytechnical University
学者数:
4.6W
论文数: 3.7W
被引数: 5.3W
引用论文

引用论文

err分享
err收藏
A Two-Phase Evolutionary Approach for Compressive Sensing Reconstruction
err2017-09-01
err42
PREAI
errZhou, Yu; Kwong, Sam; Guo, Hainan; Zhang, Xiao; Zhang, Qingfu
err分享
err收藏
Multiple Complementary Priors for Multispectral Image Compressive Sensing Reconstruction
err2024-05-01
err2
PREAI
errZha, Zhiyuan; Wen, Bihan; Yuan, Xin; Zhang, Jiachao; Zhou, Jiantao; Jiang, Xudong; Zhu, Ce
err分享
err收藏
CSformer: Bridging Convolution and Transformer for Compressive Sensing
err2023-01-01
err25
errOAAI
errYe, Dongjie; Ni, Zhangkai; Wang, Hanli; Zhang, Jian; Wang, Shiqi; Kwong, Sam
err分享
err收藏
Deep-Learned Regularization and Proximal Operator for Image Compressive Sensing
err2021-01-01
err31
errOAAI
errChen, Zan; Guo, Wenlong; Feng, Yuanjing; Li, Yongqiang; Zhao, Changchen; Ren, Yi; Shao, Ling
err分享
err收藏
Image Compressed Sensing Using Convolutional Neural Network
err2020-01-01
err254
PREAI
errShi, Wuzhen; Jiang, Feng; Liu, Shaohui; Zhao, Debin
err分享
err收藏
学者 查看更多内容