{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/dr2-net-deep-residual-reconstruction-network","title":"DR2-Net: Deep Residual Reconstruction Network for Image Compressive Sensing","arxiv_id":"1702.05743","date":"2017-02-19","proceeding":null,"authors":["Hantao Yao","Feng Dai","Dongming Zhang","Yike Ma","Shiliang Zhang","Yongdong Zhang","Qi Tian"],"abstract":"Most traditional algorithms for compressive sensing image reconstruction\nsuffer from the intensive computation. Recently, deep learning-based\nreconstruction algorithms have been reported, which dramatically reduce the\ntime complexity than iterative reconstruction algorithms. In this paper, we\npropose a novel \\textbf{D}eep \\textbf{R}esidual \\textbf{R}econstruction Network\n(DR$^{2}$-Net) to reconstruct the image from its Compressively Sensed (CS)\nmeasurement. The DR$^{2}$-Net is proposed based on two observations: 1) linear\nmapping could reconstruct a high-quality preliminary image, and 2) residual\nlearning could further improve the reconstruction quality. Accordingly,\nDR$^{2}$-Net consists of two components, \\emph{i.e.,} linear mapping network\nand residual network, respectively. Specifically, the fully-connected layer in\nneural network implements the linear mapping network. We then expand the linear\nmapping network to DR$^{2}$-Net by adding several residual learning blocks to\nenhance the preliminary image. Extensive experiments demonstrate that the\nDR$^{2}$-Net outperforms traditional iterative methods and recent deep\nlearning-based methods by large margins at measurement rates 0.01, 0.04, 0.1,\nand 0.25, respectively. The code of DR$^{2}$-Net has been released on:\nhttps://github.com/coldrainyht/caffe\\_dr2","url_abs":"http://arxiv.org/abs/1702.05743v4","url_pdf":"http://arxiv.org/pdf/1702.05743v4.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"dr2-net-deep-residual-reconstruction-network","repo_url":"https://github.com/coldrainyht/caffe_dr2","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"compressive-sensing","task_name":"Compressive Sensing"},{"task_slug":"image-reconstruction","task_name":"Image Reconstruction"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1702.05743","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}