Natural image matting with membership propagation

Weiwei Du, Kiichi Urahama

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

We present a semi-supervised technique of object extraction for natural image matting. At first, we present a novel unsupervised graph-spectral algorithm for extraction of homogeneous regions in an image. We next derive a semi-supervised scheme from this unsupervised algorithm. In our method, it is sufficient for users to draw strokes only in one of object and background regions. The semi-supervised optimization problem is solved with an iterative method where memberships are propagated from strokes to their surroundings. We suggest a guideline for placement of strokes by exploiting the same iterative solution process in the unsupervised algorithm. We project the color vectors with the linear discriminant analysis to improve the color discriminability and speed up the convergence of the iterative method. Performance of the proposed method is examined for some images and the results are compared with other methods and ground truth mattes.

Original languageEnglish
Pages (from-to)3-11
Number of pages9
JournalIPSJ Transactions on Computer Vision and Applications
Volume1
DOIs
Publication statusPublished - Dec 1 2009

Fingerprint

Iterative methods
Color
Discriminant analysis

All Science Journal Classification (ASJC) codes

  • Computer Vision and Pattern Recognition

Cite this

Natural image matting with membership propagation. / Du, Weiwei; Urahama, Kiichi.

In: IPSJ Transactions on Computer Vision and Applications, Vol. 1, 01.12.2009, p. 3-11.

Research output: Contribution to journalArticle

@article{2f6bcc5bbe2a4830853bb3d73b21db16,
title = "Natural image matting with membership propagation",
abstract = "We present a semi-supervised technique of object extraction for natural image matting. At first, we present a novel unsupervised graph-spectral algorithm for extraction of homogeneous regions in an image. We next derive a semi-supervised scheme from this unsupervised algorithm. In our method, it is sufficient for users to draw strokes only in one of object and background regions. The semi-supervised optimization problem is solved with an iterative method where memberships are propagated from strokes to their surroundings. We suggest a guideline for placement of strokes by exploiting the same iterative solution process in the unsupervised algorithm. We project the color vectors with the linear discriminant analysis to improve the color discriminability and speed up the convergence of the iterative method. Performance of the proposed method is examined for some images and the results are compared with other methods and ground truth mattes.",
author = "Weiwei Du and Kiichi Urahama",
year = "2009",
month = "12",
day = "1",
doi = "10.2197/ipsjtcva.1.3",
language = "English",
volume = "1",
pages = "3--11",
journal = "IPSJ Transactions on Computer Vision and Applications",
issn = "1882-6695",
publisher = "Information Processing Society of Japan",

}

TY - JOUR

T1 - Natural image matting with membership propagation

AU - Du, Weiwei

AU - Urahama, Kiichi

PY - 2009/12/1

Y1 - 2009/12/1

N2 - We present a semi-supervised technique of object extraction for natural image matting. At first, we present a novel unsupervised graph-spectral algorithm for extraction of homogeneous regions in an image. We next derive a semi-supervised scheme from this unsupervised algorithm. In our method, it is sufficient for users to draw strokes only in one of object and background regions. The semi-supervised optimization problem is solved with an iterative method where memberships are propagated from strokes to their surroundings. We suggest a guideline for placement of strokes by exploiting the same iterative solution process in the unsupervised algorithm. We project the color vectors with the linear discriminant analysis to improve the color discriminability and speed up the convergence of the iterative method. Performance of the proposed method is examined for some images and the results are compared with other methods and ground truth mattes.

AB - We present a semi-supervised technique of object extraction for natural image matting. At first, we present a novel unsupervised graph-spectral algorithm for extraction of homogeneous regions in an image. We next derive a semi-supervised scheme from this unsupervised algorithm. In our method, it is sufficient for users to draw strokes only in one of object and background regions. The semi-supervised optimization problem is solved with an iterative method where memberships are propagated from strokes to their surroundings. We suggest a guideline for placement of strokes by exploiting the same iterative solution process in the unsupervised algorithm. We project the color vectors with the linear discriminant analysis to improve the color discriminability and speed up the convergence of the iterative method. Performance of the proposed method is examined for some images and the results are compared with other methods and ground truth mattes.

UR - http://www.scopus.com/inward/record.url?scp=79954480735&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79954480735&partnerID=8YFLogxK

U2 - 10.2197/ipsjtcva.1.3

DO - 10.2197/ipsjtcva.1.3

M3 - Article

VL - 1

SP - 3

EP - 11

JO - IPSJ Transactions on Computer Vision and Applications

JF - IPSJ Transactions on Computer Vision and Applications

SN - 1882-6695

ER -