Paper Info
Title
K-WEB: Nonnegative dictionary learning for sparse image representations
Abstract
This paper presents a new nonnegative dictionary learning method, to decompose an input data matrix into a dictionary of nonnegative atoms, and a representation matrix with a strict ℓ0-sparsity constraint. This constraint makes each input vector representable by a limited combination of atoms. The proposed method consists of two steps which are alternatively iterated: a sparse coding and a dictionary update stage. As for the dictionary update, an original method is proposed, which we call K-WEB, as it involves the computation of k WEighted Barycenters. The so designed algorithm is shown to outperform other methods in the literature that address the same learning problem, in different applications, and both with synthetic and “real” data, i.e. coming from natural images.
Year
DOI
Venue
2013
10.1109/ICIP.2013.6738031
Image Processing
Keywords
Field
DocType
dictionaries,image coding,image representation,learning (artificial intelligence),matrix decomposition,K-WEB,dictionary update stage,input data matrix decomposition,k weighted barycenters,natural images,nonnegative atoms dictionary,nonnegative dictionary learning,representation matrix,sparse coding,sparse image representation,strict ℓ0-sparsity constraint,Dictionary learning,K-SVD,NMF,sparse representations
Pattern recognition,K-SVD,Neural coding,Computer science,Matrix (mathematics),Matrix decomposition,Theoretical computer science,Sparse image,Artificial intelligence,Iterated function,Sparse matrix,Computation
Conference
ISSN
Citations 
PageRank 
1522-4880
2
0.36
References 
Authors
2
4
Name
Order
Citations
PageRank
Marco Bevilacqua131012.14
Aline Roumy246432.54
Christine Guillemot31286104.25
Marie-Line Alberi-Morel435120.31