Presentation 2007-09-28
Shift-invariant Sparse Representations of Texture Images
Makoto NAKASHIZUKA, HIDENARI Nishiura, Yoji IIGUNI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This paper presents a shift-invariant sparse image representation for extraction of micro structures of image textures. Our image representation is based on a sparse coding which is a generative model of images and can extract salient features of images. We introduce a shift-invariant non-negative sparse representation for extraction of texture elements of images. In this paper, we employ a block coordinate relaxation algorithm which is an efficient algorithm of basis pursuit denoising for updating of weights of the shifted texture elements. We demonstrate the extraction of texture elements from images and texture image separation and interpolation via the proposed representation.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Texture image / sparse signal representation / sparse coding / unsupervised learning
Paper # SIP2007-101,SIS2007-37,SP2007-63
Date of Issue

Conference Information
Committee SIS
Conference Date 2007/9/21(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Smart Info-Media Systems (SIS)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Shift-invariant Sparse Representations of Texture Images
Sub Title (in English)
Keyword(1) Texture image
Keyword(2) sparse signal representation
Keyword(3) sparse coding
Keyword(4) unsupervised learning
1st Author's Name Makoto NAKASHIZUKA
1st Author's Affiliation Graduate School of Engineering Science, Osaka University()
2nd Author's Name HIDENARI Nishiura
2nd Author's Affiliation Graduate School of Engineering Science, Osaka University
3rd Author's Name Yoji IIGUNI
3rd Author's Affiliation Graduate School of Engineering Science, Osaka University
Date 2007-09-28
Paper # SIP2007-101,SIS2007-37,SP2007-63
Volume (vol) vol.107
Number (no) 237
Page pp.pp.-
#Pages 6
Date of Issue