CSformer : Bridging Convolution and Transformer for Compressive Sensing
Research output: Journal Publications and Reviews › RGC 21 - Publication in refereed journal › peer-review
Author(s)
Related Research Unit(s)
Detail(s)
Original language | English |
---|---|
Pages (from-to) | 2827-2842 |
Number of pages | 16 |
Journal / Publication | IEEE Transactions on Image Processing |
Volume | 32 |
Online published | 15 May 2023 |
Publication status | Published - 2023 |
Link(s)
DOI | DOI |
---|---|
Document Link | Links
|
Link to Scopus | https://www.scopus.com/record/display.uri?eid=2-s2.0-85159773152&origin=recordpage |
Permanent Link | https://scholars.cityu.edu.hk/en/publications/publication(837bc9cc-412a-4e0e-a1b7-5727f9bc9c6c).html |
Abstract
Convolutional Neural Networks (CNNs) dominate image processing but suffer from local inductive bias, which is addressed by the transformer framework with its inherent ability to capture global context through self-attention mechanisms. However, how to inherit and integrate their advantages to improve compressed sensing is still an open issue. This paper proposes CSformer, a hybrid framework to explore the representation capacity of local and global features. The proposed approach is well-designed for end-to-end compressive image sensing, composed of adaptive sampling and recovery. In the sampling module, images are measured block-by-block by the learned sampling matrix. In the reconstruction stage, the measurements are projected into an initialization stem, a CNN stem, and a transformer stem. The initialization stem mimics the traditional reconstruction of compressive sensing but generates the initial reconstruction in a learnable and efficient manner. The CNN stem and transformer stem are concurrent, simultaneously calculating fine-grained and long-range features and efficiently aggregating them. Furthermore, we explore a progressive strategy and window-based transformer block to reduce the parameters and computational complexity. The experimental results demonstrate the effectiveness of the dedicated transformer-based architecture for compressive sensing, which achieves superior performance compared to state-of-the-art methods on different datasets. Our codes is available at: https://github.com/Lineves7/CSformer. © 2023 IEEE.
Research Area(s)
- CNN, Compressive sensing, image reconstruction, transformer
Citation Format(s)
CSformer: Bridging Convolution and Transformer for Compressive Sensing. / Ye, Dongjie; Ni, Zhangkai; Wang, Hanli et al.
In: IEEE Transactions on Image Processing, Vol. 32, 2023, p. 2827-2842.
In: IEEE Transactions on Image Processing, Vol. 32, 2023, p. 2827-2842.
Research output: Journal Publications and Reviews › RGC 21 - Publication in refereed journal › peer-review