Dual-Path Attention Network for Compressed Sensing Image Reconstruction

• Published in: IEEE transactions on image processing Vol. 29; pp. 9482 - 9495
• Main Authors: Sun, Yubao, Chen, Jiwei, Liu, Qingshan, Liu, Bo, Guo, Guodong
• Format: Journal Article
• Language: English
• Published: United States IEEE 01.01.2020; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Artificial neural networks; Compressed sensing; dual-path network; Image coding; Image reconstruction; Iterative methods; Machine learning; Modules; Optimization; Periodic structures; structure-texture decomposition; Texture; texture attention
• ISSN: 1057-7149; 1941-0042; 1941-0042
• DOI: 10.1109/TIP.2020.3023629

Although deep neural network methods achieved much success in compressed sensing image reconstruction in recent years, they still have some issues, especially in preserving texture details. In this article, we propose a new dual-path attention network for compressed sensing image reconstruction, which is composed of a structure path, a texture path and a texture attention module. Motivated by the classical paradigm of image structure-texture decomposition, the structure path aims to reconstruct the dominant structure component of the original image, and the texture path targets at recovering the remaining texture details. To better bridge the information between two paths, the texture attention module is designed to deliver the useful structure information to the texture path and predict the texture region, thereby facilitating the recovery of texture details. Two paths are optimized with a unified loss function. In the testing phase, given the measurement vector of a new image, it can be well reconstructed by carrying out the well trained dual-path attention network and integrating the outputs of the structure path and the texture path. Experimental results on the SET5, SET11 and BSD68 testing datasets demonstrate that the proposed method achieves comparable or better results compared with some state-of-the-art deep learning based methods and conventional iterative optimization based methods in terms of reconstruction quality and robustness to noise.