|
Published Articles >> Table of Contents >> Abstract
Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1
pp. 764-771
A Multi-Scale Hybrid Linear Model for Lossy Image Representation
Wei Hong, University of Illinois at Urbana-Champaign
John Wright, University of Illinois at Urbana-Champaign
Kun Huang, Ohio State University
Yi Ma, University of Illinois at Urbana-Champaign
Full Article Text:
 
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/ICCV.2005.12
Send link to a friend
| Abstract |
|
This paper introduces a simple and efficient representation
for natural images. We partition an image into blocks and
treat the blocks as vectors in a high-dimensional space. We
then fit a piece-wise linear model (i.e. a union of affine
subspaces) to the vectors at each down-sampling scale. We
call this a multi-scale hybrid linear model of the image. The
hybrid and hierarchical structure of this model allows us
effectively to extract and exploit multi-modal correlations
among the imagery data at different scales. It conceptually
and computationally remedies limitations of many existing
image representation methods that are based on either
a fixed linear transformation (e.g. DCT, wavelets), an
adaptive uni-modal linear transformation (e.g. PCA), or a
multi-modal model at a single scale. We will justify both
analytically and experimentally why and how such a simple
multi-scale hybrid model is able to reduce simultaneously
the model complexity and computational cost. Despite a
small overhead for the model, our results show that this new
model gives more compact representations for a wide variety
of natural images under a wide range of signal-to-noise
ratio than many existing methods, including wavelets.
|
 Additional Information
|
Citation:
Wei Hong, John Wright, Kun Huang, Yi Ma,
"A Multi-Scale Hybrid Linear Model for Lossy Image Representation,"
iccv,
pp. 764-771,
Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1,
2005
|
|
