Compressive Sensing via Nonlocal Low-Rank Regularization. Dong, W., Shi, G., Li, X., Ma, Y., & Huang, F. IEEE Transactions on Image Processing, 23(8):3618--3632, August, 2014. 00001doi abstract bibtex Sparsity has been widely exploited for exact reconstruction of a signal from a small number of random measurements. Recent advances have suggested that structured or group sparsity often leads to more powerful signal reconstruction techniques in various compressed sensing (CS) studies. In this paper, we propose a nonlocal low-rank regularization (NLR) approach toward exploiting structured sparsity and explore its application into CS of both photographic and MRI images. We also propose the use of a nonconvex log det ( X) as a smooth surrogate function for the rank instead of the convex nuclear norm and justify the benefit of such a strategy using extensive experiments. To further improve the computational efficiency of the proposed algorithm, we have developed a fast implementation using the alternative direction multiplier method technique. Experimental results have shown that the proposed NLR-CS algorithm can significantly outperform existing state-of-the-art CS techniques for image recovery.
@article{ dong_compressive_2014,
title = {Compressive {Sensing} via {Nonlocal} {Low}-{Rank} {Regularization}},
volume = {23},
issn = {1057-7149},
doi = {10.1109/TIP.2014.2329449},
abstract = {Sparsity has been widely exploited for exact reconstruction of a signal from a small number of random measurements. Recent advances have suggested that structured or group sparsity often leads to more powerful signal reconstruction techniques in various compressed sensing (CS) studies. In this paper, we propose a nonlocal low-rank regularization (NLR) approach toward exploiting structured sparsity and explore its application into CS of both photographic and MRI images. We also propose the use of a nonconvex log det ( X) as a smooth surrogate function for the rank instead of the convex nuclear norm and justify the benefit of such a strategy using extensive experiments. To further improve the computational efficiency of the proposed algorithm, we have developed a fast implementation using the alternative direction multiplier method technique. Experimental results have shown that the proposed NLR-CS algorithm can significantly outperform existing state-of-the-art CS techniques for image recovery.},
number = {8},
journal = {IEEE Transactions on Image Processing},
author = {Dong, Weisheng and Shi, Guangming and Li, Xin and Ma, Yi and Huang, Feng},
month = {August},
year = {2014},
note = {00001},
pages = {3618--3632}
}
Downloads: 0
{"_id":"pb3sPnL5cLdqrbFWJ","bibbaseid":"dong-shi-li-ma-huang-compressivesensingvianonlocallowrankregularization-2014","downloads":0,"creationDate":"2015-09-03T07:18:48.168Z","title":"Compressive Sensing via Nonlocal Low-Rank Regularization","author_short":["Dong, W.","Shi, G.","Li, X.","Ma, Y.","Huang, F."],"year":2014,"bibtype":"article","biburl":"http://bibbase.org/zotero/fred.qi","bibdata":{"abstract":"Sparsity has been widely exploited for exact reconstruction of a signal from a small number of random measurements. Recent advances have suggested that structured or group sparsity often leads to more powerful signal reconstruction techniques in various compressed sensing (CS) studies. In this paper, we propose a nonlocal low-rank regularization (NLR) approach toward exploiting structured sparsity and explore its application into CS of both photographic and MRI images. We also propose the use of a nonconvex log det ( X) as a smooth surrogate function for the rank instead of the convex nuclear norm and justify the benefit of such a strategy using extensive experiments. To further improve the computational efficiency of the proposed algorithm, we have developed a fast implementation using the alternative direction multiplier method technique. Experimental results have shown that the proposed NLR-CS algorithm can significantly outperform existing state-of-the-art CS techniques for image recovery.","author":["Dong, Weisheng","Shi, Guangming","Li, Xin","Ma, Yi","Huang, Feng"],"author_short":["Dong, W.","Shi, G.","Li, X.","Ma, Y.","Huang, F."],"bibtex":"@article{ dong_compressive_2014,\n title = {Compressive {Sensing} via {Nonlocal} {Low}-{Rank} {Regularization}},\n volume = {23},\n issn = {1057-7149},\n doi = {10.1109/TIP.2014.2329449},\n abstract = {Sparsity has been widely exploited for exact reconstruction of a signal from a small number of random measurements. Recent advances have suggested that structured or group sparsity often leads to more powerful signal reconstruction techniques in various compressed sensing (CS) studies. In this paper, we propose a nonlocal low-rank regularization (NLR) approach toward exploiting structured sparsity and explore its application into CS of both photographic and MRI images. We also propose the use of a nonconvex log det ( X) as a smooth surrogate function for the rank instead of the convex nuclear norm and justify the benefit of such a strategy using extensive experiments. To further improve the computational efficiency of the proposed algorithm, we have developed a fast implementation using the alternative direction multiplier method technique. Experimental results have shown that the proposed NLR-CS algorithm can significantly outperform existing state-of-the-art CS techniques for image recovery.},\n number = {8},\n journal = {IEEE Transactions on Image Processing},\n author = {Dong, Weisheng and Shi, Guangming and Li, Xin and Ma, Yi and Huang, Feng},\n month = {August},\n year = {2014},\n note = {00001},\n pages = {3618--3632}\n}","bibtype":"article","doi":"10.1109/TIP.2014.2329449","id":"dong_compressive_2014","issn":"1057-7149","journal":"IEEE Transactions on Image Processing","key":"dong_compressive_2014","month":"August","note":"00001","number":"8","pages":"3618--3632","title":"Compressive Sensing via Nonlocal Low-Rank Regularization","type":"article","volume":"23","year":"2014","bibbaseid":"dong-shi-li-ma-huang-compressivesensingvianonlocallowrankregularization-2014","role":"author","urls":{},"downloads":0},"search_terms":["compressive","sensing","via","nonlocal","low","rank","regularization","dong","shi","li","ma","huang"],"keywords":[],"authorIDs":[],"dataSources":["y8en6y5RHukfeLuPH"]}