DocumentCode :
2034568
Title :
Multiscale Sparse Image Representationwith Learned Dictionaries
Author :
Mairal, Julien ; Sapiro, Guillermo ; Elad, Michael
Author_Institution :
Univ. of Minnesota, Minneapolis
Volume :
3
fYear :
2007
fDate :
Sept. 16 2007-Oct. 19 2007
Abstract :
This paper introduces a new framework for learning multiscale sparse representations of natural images with overcomplete dictionaries. Our work extends the K-SVD algorithm [1], which learns sparse single-scale dictionaries for natural images. Recent work has shown that the K-SVD can lead to state-of-the-art image restoration results [2, 3]. We show that these are further improved with a multi-scale approach, based on a Quadtree decomposition. Our framework provides an alternative to multiscale pre-defined dictionaries such as wavelets, curvelets, and contourlets, with dictionaries optimized for the data and application instead of pre-modelled ones.
Keywords :
image denoising; image representation; image restoration; quadtrees; singular value decomposition; sparse matrices; K-SVD algorithm; Quadtree decomposition; image denoising; image restoration; multiscale sparse image representation; natural images; overcomplete dictionaries; sparse single-scale dictionary learning; Approximation algorithms; Dictionaries; Equations; Gaussian noise; Image processing; Image representation; Image restoration; Matching pursuit algorithms; Minimization methods; Noise reduction; Denoising; Image Restoration; Multiscale; Sparsity;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Image Processing, 2007. ICIP 2007. IEEE International Conference on
Conference_Location :
San Antonio, TX
ISSN :
1522-4880
Print_ISBN :
978-1-4244-1437-6
Electronic_ISBN :
1522-4880
Type :
conf
DOI :
10.1109/ICIP.2007.4379257
Filename :
4379257
Link To Document :
بازگشت