Introducing a weighted non-negative matrix factorization for image classification

@article{citeulike:423477, abstract = {[Non-negative matrix factorization (NMF), principal component analysis, color histogram classification, probabilistic approach] Non-negative matrix factorization (NMF) technique has been recently proposed for dimensionality reduction. NMF is capable to produce region or part based representations of objects and images. Also, a direct modification of NMF, the weighted non-negative matrix factorization (WNMF) has also been introduced to improve the NMF capabilities of representing positive local data (as color histograms). A comparison between NMF, WNMF and the well-known principal component analysis (PCA) in the context of image patch classification has been carried out and it is claimed that all these three techniques can be combined in a common and unique classifier. This contribution is an extension of a previous study and we introduce the use of the WNMF as well as a probabilistic approach to compare all the three techniques noticing a great improvement in the final recognition results.}, author = {Guillamet, D. and Vitria, J. and Schiele, B. }, citeulike-article-id = {423477}, doi = {10.1016/S0167-8655(03)00089-8}, journal = {Pattern Recognition Letters}, keywords = {nmf}, month = {October}, number = {14}, pages = {2447--2454}, posted-at = {2005-12-06 16:16:33}, priority = {0}, title = {Introducing a weighted non-negative matrix factorization for image classification}, url = {http://dx.doi.org/10.1016/S0167-8655(03)00089-8}, volume = {24}, year = {2003} }

TY - JOUR ID - citeulike:423477 L3 - citeulike-article-id:423477 TI - Introducing a weighted non-negative matrix factorization for image classification JF - Pattern Recognition Letters VL - 24 IS - 14 SP - 2447 EP - 2454 N2 - [Non-negative matrix factorization (NMF), principal component analysis, color histogram classification, probabilistic approach] Non-negative matrix factorization (NMF) technique has been recently proposed for dimensionality reduction. NMF is capable to produce region or part based representations of objects and images. Also, a direct modification of NMF, the weighted non-negative matrix factorization (WNMF) has also been introduced to improve the NMF capabilities of representing positive local data (as color histograms). A comparison between NMF, WNMF and the well-known principal component analysis (PCA) in the context of image patch classification has been carried out and it is claimed that all these three techniques can be combined in a common and unique classifier. This contribution is an extension of a previous study and we introduce the use of the WNMF as well as a probabilistic approach to compare all the three techniques noticing a great improvement in the final recognition results. KW - nmf AU - Guillamet, D AU - Vitria, J AU - Schiele, B PY - 2003/10// UR - http://dx.doi.org/10.1016/S0167-8655(03)00089-8 ER -