Paper Info
Title
Farness preserving Non-negative matrix factorization
Abstract
Dramatic growth in the volume of data made a compact and informative representation of the data highly demanded in computer vision, information retrieval, and pattern recognition. Non-negative Matrix Factorization (NMF) is used widely to provide parts-based representations by factorizing the data matrix into non-negative matrix factors. Since non-negativity constraint is not sufficient to achieve robust results, variants of NMF have been introduced to exploit the geometry of the data space. While these variants considered the local invariance based on the manifold assumption, we propose Farness preserving Non-negative Matrix Factorization (FNMF) to exploits the geometry of the data space by considering non-local invariance which is applicable to any data structure. FNMF adds a new constraint to enforce the far points (i.e., non-neighbors) in original space to stay far in the new space. Experiments on different kinds of data (e.g., Multimedia, Earth Observation) demonstrate that FNMF outperforms the other variants of NMF.
Year
DOI
Venue
2014
10.1109/ICIP.2014.7025611
Image Processing
Keywords
Field
DocType
computational geometry,data structures,matrix decomposition,FNMF,compact data representation,computer vision,data matrix,data space geometry exploitation,data structure,farness preserving nonnegative matrix factorization,information retrieval,informative data representation,parts-based representations,pattern recognition,Clustering,Farness Preserving,Non-negative Matrix Factorization
Data structure,Essential matrix,Invariant (physics),Matrix (mathematics),Computer science,Matrix decomposition,Theoretical computer science,Non-negative matrix factorization,Cluster analysis,Manifold
Conference
ISSN
Citations 
PageRank 
1522-4880
4
0.47
References 
Authors
3
4
Name
Order
Citations
PageRank
Mohammadreza Babaee1736.44
Reza Bahmanyar2355.32
Gerhard Rigoll32788268.87
Mihai Datcu4893111.62