Robust Face Recognition via Sparse Representation. Wright, J., Yang, A. Y., Ganesh, A., Sastry, S. S., & Ma, Y. IEEE Transactions on Pattern Analysis and Machine Intelligence, 31(2):210–227, February, 2009. Conference Name: IEEE Transactions on Pattern Analysis and Machine Intelligence rate: 5doi abstract bibtex We consider the problem of automatically recognizing human faces from frontal views with varying expression and illumination, as well as occlusion and disguise. We cast the recognition problem as one of classifying among multiple linear regression models and argue that new theory from sparse signal representation offers the key to addressing this problem. Based on a sparse representation computed by C1-minimization, we propose a general classification algorithm for (image-based) object recognition. This new framework provides new insights into two crucial issues in face recognition: feature extraction and robustness to occlusion. For feature extraction, we show that if sparsity in the recognition problem is properly harnessed, the choice of features is no longer critical. What is critical, however, is whether the number of features is sufficiently large and whether the sparse representation is correctly computed. Unconventional features such as downsampled images and random projections perform just as well as conventional features such as eigenfaces and Laplacianfaces, as long as the dimension of the feature space surpasses certain threshold, predicted by the theory of sparse representation. This framework can handle errors due to occlusion and corruption uniformly by exploiting the fact that these errors are often sparse with respect to the standard (pixel) basis. The theory of sparse representation helps predict how much occlusion the recognition algorithm can handle and how to choose the training images to maximize robustness to occlusion. We conduct extensive experiments on publicly available databases to verify the efficacy of the proposed algorithm and corroborate the above claims.
@article{wright_robust_2009,
title = {Robust {Face} {Recognition} via {Sparse} {Representation}},
volume = {31},
issn = {1939-3539},
doi = {10.1109/TPAMI.2008.79},
abstract = {We consider the problem of automatically recognizing human faces from frontal views with varying expression and illumination, as well as occlusion and disguise. We cast the recognition problem as one of classifying among multiple linear regression models and argue that new theory from sparse signal representation offers the key to addressing this problem. Based on a sparse representation computed by C1-minimization, we propose a general classification algorithm for (image-based) object recognition. This new framework provides new insights into two crucial issues in face recognition: feature extraction and robustness to occlusion. For feature extraction, we show that if sparsity in the recognition problem is properly harnessed, the choice of features is no longer critical. What is critical, however, is whether the number of features is sufficiently large and whether the sparse representation is correctly computed. Unconventional features such as downsampled images and random projections perform just as well as conventional features such as eigenfaces and Laplacianfaces, as long as the dimension of the feature space surpasses certain threshold, predicted by the theory of sparse representation. This framework can handle errors due to occlusion and corruption uniformly by exploiting the fact that these errors are often sparse with respect to the standard (pixel) basis. The theory of sparse representation helps predict how much occlusion the recognition algorithm can handle and how to choose the training images to maximize robustness to occlusion. We conduct extensive experiments on publicly available databases to verify the efficacy of the proposed algorithm and corroborate the above claims.},
language = {en},
number = {2},
journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence},
author = {Wright, John and Yang, Allen Y. and Ganesh, Arvind and Sastry, S. Shankar and Ma, Yi},
month = feb,
year = {2009},
note = {Conference Name: IEEE Transactions on Pattern Analysis and Machine Intelligence
rate: 5},
keywords = {\#Recognition{\textgreater}Face, \#Sparse, /unread, Classification algorithms, Classifier design and evaluation, Face and gesture recognition, Face recognition, Feature evaluation and selection, Feature extraction, Humans, Image recognition, Lighting, Linear regression, Object recognition, Occlusion, Outlier rejection, Robustness, Signal representations, Spare representation, {\textbackslash}ell{\textasciicircum}1–minimization, compressed sensing, feature extraction, occlusion and corruption, sparse representation, validation and outlier rejection., ⭐⭐⭐⭐⭐},
pages = {210--227},
}
Downloads: 0
{"_id":"69u9DF3SGo7S2zeQ4","bibbaseid":"wright-yang-ganesh-sastry-ma-robustfacerecognitionviasparserepresentation-2009","downloads":0,"creationDate":"2015-09-03T07:18:47.077Z","title":"Robust Face Recognition via Sparse Representation","author_short":["Wright, J.","Yang, A. Y.","Ganesh, A.","Sastry, S. S.","Ma, Y."],"year":2009,"bibtype":"article","biburl":"https://bibbase.org/zotero/zzhenry2012","bibdata":{"bibtype":"article","type":"article","title":"Robust Face Recognition via Sparse Representation","volume":"31","issn":"1939-3539","doi":"10.1109/TPAMI.2008.79","abstract":"We consider the problem of automatically recognizing human faces from frontal views with varying expression and illumination, as well as occlusion and disguise. We cast the recognition problem as one of classifying among multiple linear regression models and argue that new theory from sparse signal representation offers the key to addressing this problem. Based on a sparse representation computed by C1-minimization, we propose a general classification algorithm for (image-based) object recognition. This new framework provides new insights into two crucial issues in face recognition: feature extraction and robustness to occlusion. For feature extraction, we show that if sparsity in the recognition problem is properly harnessed, the choice of features is no longer critical. What is critical, however, is whether the number of features is sufficiently large and whether the sparse representation is correctly computed. Unconventional features such as downsampled images and random projections perform just as well as conventional features such as eigenfaces and Laplacianfaces, as long as the dimension of the feature space surpasses certain threshold, predicted by the theory of sparse representation. This framework can handle errors due to occlusion and corruption uniformly by exploiting the fact that these errors are often sparse with respect to the standard (pixel) basis. The theory of sparse representation helps predict how much occlusion the recognition algorithm can handle and how to choose the training images to maximize robustness to occlusion. We conduct extensive experiments on publicly available databases to verify the efficacy of the proposed algorithm and corroborate the above claims.","language":"en","number":"2","journal":"IEEE Transactions on Pattern Analysis and Machine Intelligence","author":[{"propositions":[],"lastnames":["Wright"],"firstnames":["John"],"suffixes":[]},{"propositions":[],"lastnames":["Yang"],"firstnames":["Allen","Y."],"suffixes":[]},{"propositions":[],"lastnames":["Ganesh"],"firstnames":["Arvind"],"suffixes":[]},{"propositions":[],"lastnames":["Sastry"],"firstnames":["S.","Shankar"],"suffixes":[]},{"propositions":[],"lastnames":["Ma"],"firstnames":["Yi"],"suffixes":[]}],"month":"February","year":"2009","note":"Conference Name: IEEE Transactions on Pattern Analysis and Machine Intelligence rate: 5","keywords":"#Recognition\\textgreaterFace, #Sparse, /unread, Classification algorithms, Classifier design and evaluation, Face and gesture recognition, Face recognition, Feature evaluation and selection, Feature extraction, Humans, Image recognition, Lighting, Linear regression, Object recognition, Occlusion, Outlier rejection, Robustness, Signal representations, Spare representation, \\ell\\textasciicircum1–minimization, compressed sensing, feature extraction, occlusion and corruption, sparse representation, validation and outlier rejection., ⭐⭐⭐⭐⭐","pages":"210–227","bibtex":"@article{wright_robust_2009,\n\ttitle = {Robust {Face} {Recognition} via {Sparse} {Representation}},\n\tvolume = {31},\n\tissn = {1939-3539},\n\tdoi = {10.1109/TPAMI.2008.79},\n\tabstract = {We consider the problem of automatically recognizing human faces from frontal views with varying expression and illumination, as well as occlusion and disguise. We cast the recognition problem as one of classifying among multiple linear regression models and argue that new theory from sparse signal representation offers the key to addressing this problem. Based on a sparse representation computed by C1-minimization, we propose a general classification algorithm for (image-based) object recognition. This new framework provides new insights into two crucial issues in face recognition: feature extraction and robustness to occlusion. For feature extraction, we show that if sparsity in the recognition problem is properly harnessed, the choice of features is no longer critical. What is critical, however, is whether the number of features is sufficiently large and whether the sparse representation is correctly computed. Unconventional features such as downsampled images and random projections perform just as well as conventional features such as eigenfaces and Laplacianfaces, as long as the dimension of the feature space surpasses certain threshold, predicted by the theory of sparse representation. This framework can handle errors due to occlusion and corruption uniformly by exploiting the fact that these errors are often sparse with respect to the standard (pixel) basis. The theory of sparse representation helps predict how much occlusion the recognition algorithm can handle and how to choose the training images to maximize robustness to occlusion. We conduct extensive experiments on publicly available databases to verify the efficacy of the proposed algorithm and corroborate the above claims.},\n\tlanguage = {en},\n\tnumber = {2},\n\tjournal = {IEEE Transactions on Pattern Analysis and Machine Intelligence},\n\tauthor = {Wright, John and Yang, Allen Y. and Ganesh, Arvind and Sastry, S. Shankar and Ma, Yi},\n\tmonth = feb,\n\tyear = {2009},\n\tnote = {Conference Name: IEEE Transactions on Pattern Analysis and Machine Intelligence\nrate: 5},\n\tkeywords = {\\#Recognition{\\textgreater}Face, \\#Sparse, /unread, Classification algorithms, Classifier design and evaluation, Face and gesture recognition, Face recognition, Feature evaluation and selection, Feature extraction, Humans, Image recognition, Lighting, Linear regression, Object recognition, Occlusion, Outlier rejection, Robustness, Signal representations, Spare representation, {\\textbackslash}ell{\\textasciicircum}1–minimization, compressed sensing, feature extraction, occlusion and corruption, sparse representation, validation and outlier rejection., ⭐⭐⭐⭐⭐},\n\tpages = {210--227},\n}\n\n\n\n","author_short":["Wright, J.","Yang, A. Y.","Ganesh, A.","Sastry, S. S.","Ma, Y."],"key":"wright_robust_2009","id":"wright_robust_2009","bibbaseid":"wright-yang-ganesh-sastry-ma-robustfacerecognitionviasparserepresentation-2009","role":"author","urls":{},"keyword":["#Recognition\\textgreaterFace","#Sparse","/unread","Classification algorithms","Classifier design and evaluation","Face and gesture recognition","Face recognition","Feature evaluation and selection","Feature extraction","Humans","Image recognition","Lighting","Linear regression","Object recognition","Occlusion","Outlier rejection","Robustness","Signal representations","Spare representation","\\ell\\textasciicircum1–minimization","compressed sensing","feature extraction","occlusion and corruption","sparse representation","validation and outlier rejection.","⭐⭐⭐⭐⭐"],"metadata":{"authorlinks":{}},"downloads":0,"html":""},"search_terms":["robust","face","recognition","via","sparse","representation","wright","yang","ganesh","sastry","ma"],"keywords":["#recognition\\textgreaterface","#sparse","/unread","classification algorithms","classifier design and evaluation","face and gesture recognition","face recognition","feature evaluation and selection","feature extraction","humans","image recognition","lighting","linear regression","object recognition","occlusion","outlier rejection","robustness","signal representations","spare representation","\\ell\\textasciicircum1–minimization","compressed sensing","feature extraction","occlusion and corruption","sparse representation","validation and outlier rejection.","⭐⭐⭐⭐⭐"],"authorIDs":[],"dataSources":["9cexBw6hrwgyZphZZ","nZHrFJKyxKKDaWYM8"]}