Matches in SemOpenAlex for { <https://semopenalex.org/work/W13755666> ?p ?o ?g. }
Showing items 1 to 100 of
100
with 100 items per page.
- W13755666 abstract "AbstractWe present a novel Discriminant Locality Preserving Projections (DLPP) algorithm named Collaborative Discrim-inant Locality Preserving Projection (CDLPP). In our algorithm, the discriminating power of DLPP are furtherexploited from two aspects. On the one hand, the global optimum of class scattering is guaranteed via using thebetween-class scatter matrix to replace the original denominator of DLPP. On the other hand, motivated by the sparserepresentation and collaborative representation, a L 2 -norm constraint is imposed to the projections to discover thecollaborations of dimensions in the sample space. We apply our algorithm to face recognition. Three popular facedatabases, namely AR, ORL and LFW-a, are employed for evaluating the performance of CDLPP. Extensive experi-mental results demonstrate that CDLPP significantly improve the discriminating power of DLPP and outperform thestate-of-the-arts.Keywords: Discriminant Locality Preserving Projections, Face recognition, Dimensionality reduction, Featureextraction, Collaborative representation1. IntroductionSubspace learning is a useful technique in computer vision, pattern recognition and machine learning, particularlyfor solving the dimensionality reduction, feature selection, feature extraction and face recognition tasks. Subspacelearning aims to learn a specific subspace of the original sample space, which has some particular desired properties.This topic has been studied for decades and many impressive algorithms have been proposed. The representativesubspace learning algorithms include Principle Component Analysis (PCA) [1, 2, 3], Linear Discriminant Analysis(LDA) [4], Non-negative Matrix factorization (NMF) [5, 6], Independent Component Analysis (ICA) [7], LocalityPreserving Projections (LPP) and so on. In face recognition, subspace learning is also known as appearance-basedface recognition. For example, PCA is known as Eigenfaces, LDA is known as Fisherfaces and LPP is known asLaplacianfaces.Some recent studies show the high-dimensional samples may reside on low-dimensional manifolds [8, 9] and suchmanifold structures are essential for data clustering and classification [10, 11]. The manifold-based subspace learningalgorithms may start from Locality Preserving Projections (LPP). LPP constructs an adjacency matrix to weightingthe distance between each two sample points for learning a projection which can preserve the local manifold structuresof data. The weight between two nearby points is much greater than the weight between two remote points. So if twopoints are close in the original space then they will be close in the learned subspace as well. However, the conventionalLPP only takes the manifold information into consideration. Many researchers make e orts to improve LPP fromdi erent perspectives. Discriminant Locality Preserving Projections (DLPP) [12, 13, 14] is deemed as one of the mostsuccessful extension of LPP. It improves the discriminating power of LPP via simultaneously maximizing the distancebetween each two nearby classes and minimizing the original LPP objective. Orthogonal Laplacianfaces (OLPP) [15]imposes an orthogonality constraint to LPP to ensure that the learned projections are mutually orthogonal. Parametric" @default.
- W13755666 created "2016-06-24" @default.
- W13755666 creator A5052560070 @default.
- W13755666 creator A5089740855 @default.
- W13755666 date "2013-12-28" @default.
- W13755666 modified "2023-09-26" @default.
- W13755666 title "Collaborative Discriminant Locality Preserving Projections With its Application to Face Recognition" @default.
- W13755666 cites W1509928947 @default.
- W13755666 cites W1591385104 @default.
- W13755666 cites W1902027874 @default.
- W13755666 cites W1975900269 @default.
- W13755666 cites W2001141328 @default.
- W13755666 cites W2052800689 @default.
- W13755666 cites W2053186076 @default.
- W13755666 cites W2059881433 @default.
- W13755666 cites W2069959554 @default.
- W13755666 cites W2070127246 @default.
- W13755666 cites W2072026719 @default.
- W13755666 cites W2095469431 @default.
- W13755666 cites W2097308346 @default.
- W13755666 cites W2097486709 @default.
- W13755666 cites W2103250033 @default.
- W13755666 cites W2103560185 @default.
- W13755666 cites W2104294146 @default.
- W13755666 cites W2108119513 @default.
- W13755666 cites W2117553576 @default.
- W13755666 cites W2121647436 @default.
- W13755666 cites W2123115309 @default.
- W13755666 cites W2129812935 @default.
- W13755666 cites W2132467081 @default.
- W13755666 cites W2138451337 @default.
- W13755666 cites W2146474141 @default.
- W13755666 cites W2163398148 @default.
- W13755666 cites W2163808566 @default.
- W13755666 hasPublicationYear "2013" @default.
- W13755666 type Work @default.
- W13755666 sameAs 13755666 @default.
- W13755666 citedByCount "0" @default.
- W13755666 crossrefType "posted-content" @default.
- W13755666 hasAuthorship W13755666A5052560070 @default.
- W13755666 hasAuthorship W13755666A5089740855 @default.
- W13755666 hasConcept C11413529 @default.
- W13755666 hasConcept C138885662 @default.
- W13755666 hasConcept C153180895 @default.
- W13755666 hasConcept C154945302 @default.
- W13755666 hasConcept C27438332 @default.
- W13755666 hasConcept C2779808786 @default.
- W13755666 hasConcept C31510193 @default.
- W13755666 hasConcept C32834561 @default.
- W13755666 hasConcept C33923547 @default.
- W13755666 hasConcept C41008148 @default.
- W13755666 hasConcept C41895202 @default.
- W13755666 hasConcept C52622490 @default.
- W13755666 hasConcept C57493831 @default.
- W13755666 hasConcept C69738355 @default.
- W13755666 hasConcept C70518039 @default.
- W13755666 hasConcept C78397625 @default.
- W13755666 hasConceptScore W13755666C11413529 @default.
- W13755666 hasConceptScore W13755666C138885662 @default.
- W13755666 hasConceptScore W13755666C153180895 @default.
- W13755666 hasConceptScore W13755666C154945302 @default.
- W13755666 hasConceptScore W13755666C27438332 @default.
- W13755666 hasConceptScore W13755666C2779808786 @default.
- W13755666 hasConceptScore W13755666C31510193 @default.
- W13755666 hasConceptScore W13755666C32834561 @default.
- W13755666 hasConceptScore W13755666C33923547 @default.
- W13755666 hasConceptScore W13755666C41008148 @default.
- W13755666 hasConceptScore W13755666C41895202 @default.
- W13755666 hasConceptScore W13755666C52622490 @default.
- W13755666 hasConceptScore W13755666C57493831 @default.
- W13755666 hasConceptScore W13755666C69738355 @default.
- W13755666 hasConceptScore W13755666C70518039 @default.
- W13755666 hasConceptScore W13755666C78397625 @default.
- W13755666 hasLocation W137556661 @default.
- W13755666 hasOpenAccess W13755666 @default.
- W13755666 hasPrimaryLocation W137556661 @default.
- W13755666 hasRelatedWork W1227844046 @default.
- W13755666 hasRelatedWork W1989079133 @default.
- W13755666 hasRelatedWork W199830248 @default.
- W13755666 hasRelatedWork W2024194293 @default.
- W13755666 hasRelatedWork W2030214912 @default.
- W13755666 hasRelatedWork W2035451505 @default.
- W13755666 hasRelatedWork W2046934501 @default.
- W13755666 hasRelatedWork W2055551805 @default.
- W13755666 hasRelatedWork W2095459639 @default.
- W13755666 hasRelatedWork W2099173991 @default.
- W13755666 hasRelatedWork W2107772706 @default.
- W13755666 hasRelatedWork W2123115309 @default.
- W13755666 hasRelatedWork W2129853012 @default.
- W13755666 hasRelatedWork W2143079421 @default.
- W13755666 hasRelatedWork W2156028522 @default.
- W13755666 hasRelatedWork W2903436285 @default.
- W13755666 hasRelatedWork W30106976 @default.
- W13755666 hasRelatedWork W3126860699 @default.
- W13755666 hasRelatedWork W2343820385 @default.
- W13755666 hasRelatedWork W3147086955 @default.
- W13755666 isParatext "false" @default.
- W13755666 isRetracted "false" @default.
- W13755666 magId "13755666" @default.
- W13755666 workType "article" @default.