Efficient image representation by anisotropic refinement in matching pursuit

Authors:
P. Vandergheynst;P. Frossard
Affiliations:
Signal Process. Lab., Swiss Fed. Inst. of Technol., Lausanne, Switzerland;-
Venue:
ICASSP '01 Proceedings of the Acoustics, Speech, and Signal Processing, 2001. on IEEE International Conference - Volume 03
Year:
2001

Citing 0
Cited 9

An efficient interpolation algorithm on anisotropic grids for functions with jump discontinuities in 2-D

Applied Numerical Mathematics
Image compression using an edge adapted redundant dictionary and wavelets

Signal Processing - Sparse approximations in signal and image processing
A shadow elimination approach in video-surveillance context

Pattern Recognition Letters
Hybrid video coding based on bidimensional matching pursuit

EURASIP Journal on Applied Signal Processing
An efficient interpolation algorithm on anisotropic grids for functions with jump discontinuities in 2-D

Applied Numerical Mathematics
Geometric video approximation using weighted matching pursuit

IEEE Transactions on Image Processing
Matching-pursuits dissimilarity measure for shape-based comparison and classification of high-dimensional data

IEEE Transactions on Fuzzy Systems
A membrane algorithm with quantum-inspired subalgorithms and its application to image processing

Natural Computing: an international journal
Using dynamic multi-swarm particle swarm optimizer to improve the image sparse decomposition based on matching pursuit

ICIC'13 Proceedings of the 9th international conference on Intelligent Computing Theories and Technology

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper presents a new image representation method based on anisotropic refinement. It has been shown that wavelets are not optimal to code 2-D objects which need true 2-D dictionaries for efficient approximation. We propose to use rotations and anisotropic scaling to build a real bi-dimensional dictionary. Matching pursuit then stands as a natural candidate to provide an image representation with an anisotropic refinement scheme. It basically decomposes the image as a series of basis functions weighted by their respective coefficients. Even if the basis functions can a priori take any form, bi-dimensional dictionaries are almost exclusively composed of two-dimensional Gabor functions. We present here a new dictionary design by introducing orientation and anisotropic refinement of a Gaussian generating function. The new dictionary permits to efficiently code 2-D objects and more particularly oriented contours. It is shown to clearly outperform common nonoriented Gabor dictionaries.