Interactive cosegmentation using global and local energy optimization

Xingping Dong, Jianbing Shen, Ling Shao, Ming Hsuan Yang

Research output: Contribution to journalArticle

37 Citations (Scopus)

Abstract

We propose a novel interactive cosegmentation method using global and local energy optimization. The global energy includes two terms: 1) the global scribbled energy and 2) the interimage energy. The first one utilizes the user scribbles to build the Gaussian mixture model and improve the cosegmentation performance. The second one is a global constraint, which attempts to match the histograms of common objects. To minimize the local energy, we apply the spline regression to learn the smoothness in a local neighborhood. This energy optimization can be converted into a constrained quadratic programming problem. To reduce the computational complexity, we propose an iterative optimization algorithm to decompose this optimization problem into several subproblems. The experimental results show that our method outperforms the state-of-the-art unsupervised cosegmentation and interactive cosegmentation methods on the iCoseg and MSRC benchmark data sets.

Original languageEnglish
Article number7156153
Pages (from-to)3966-3977
Number of pages12
JournalIEEE Transactions on Image Processing
Volume24
Issue number11
DOIs
Publication statusPublished - 2015 Nov 1

Fingerprint

Quadratic programming
Splines
Computational complexity

All Science Journal Classification (ASJC) codes

  • Software
  • Computer Graphics and Computer-Aided Design

Cite this

Dong, Xingping ; Shen, Jianbing ; Shao, Ling ; Yang, Ming Hsuan. / Interactive cosegmentation using global and local energy optimization. In: IEEE Transactions on Image Processing. 2015 ; Vol. 24, No. 11. pp. 3966-3977.
@article{70f5ae31787c4aecafc9f8deeb543de9,
title = "Interactive cosegmentation using global and local energy optimization",
abstract = "We propose a novel interactive cosegmentation method using global and local energy optimization. The global energy includes two terms: 1) the global scribbled energy and 2) the interimage energy. The first one utilizes the user scribbles to build the Gaussian mixture model and improve the cosegmentation performance. The second one is a global constraint, which attempts to match the histograms of common objects. To minimize the local energy, we apply the spline regression to learn the smoothness in a local neighborhood. This energy optimization can be converted into a constrained quadratic programming problem. To reduce the computational complexity, we propose an iterative optimization algorithm to decompose this optimization problem into several subproblems. The experimental results show that our method outperforms the state-of-the-art unsupervised cosegmentation and interactive cosegmentation methods on the iCoseg and MSRC benchmark data sets.",
author = "Xingping Dong and Jianbing Shen and Ling Shao and Yang, {Ming Hsuan}",
year = "2015",
month = "11",
day = "1",
doi = "10.1109/TIP.2015.2456636",
language = "English",
volume = "24",
pages = "3966--3977",
journal = "IEEE Transactions on Image Processing",
issn = "1057-7149",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "11",

}

Interactive cosegmentation using global and local energy optimization. / Dong, Xingping; Shen, Jianbing; Shao, Ling; Yang, Ming Hsuan.

In: IEEE Transactions on Image Processing, Vol. 24, No. 11, 7156153, 01.11.2015, p. 3966-3977.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Interactive cosegmentation using global and local energy optimization

AU - Dong, Xingping

AU - Shen, Jianbing

AU - Shao, Ling

AU - Yang, Ming Hsuan

PY - 2015/11/1

Y1 - 2015/11/1

N2 - We propose a novel interactive cosegmentation method using global and local energy optimization. The global energy includes two terms: 1) the global scribbled energy and 2) the interimage energy. The first one utilizes the user scribbles to build the Gaussian mixture model and improve the cosegmentation performance. The second one is a global constraint, which attempts to match the histograms of common objects. To minimize the local energy, we apply the spline regression to learn the smoothness in a local neighborhood. This energy optimization can be converted into a constrained quadratic programming problem. To reduce the computational complexity, we propose an iterative optimization algorithm to decompose this optimization problem into several subproblems. The experimental results show that our method outperforms the state-of-the-art unsupervised cosegmentation and interactive cosegmentation methods on the iCoseg and MSRC benchmark data sets.

AB - We propose a novel interactive cosegmentation method using global and local energy optimization. The global energy includes two terms: 1) the global scribbled energy and 2) the interimage energy. The first one utilizes the user scribbles to build the Gaussian mixture model and improve the cosegmentation performance. The second one is a global constraint, which attempts to match the histograms of common objects. To minimize the local energy, we apply the spline regression to learn the smoothness in a local neighborhood. This energy optimization can be converted into a constrained quadratic programming problem. To reduce the computational complexity, we propose an iterative optimization algorithm to decompose this optimization problem into several subproblems. The experimental results show that our method outperforms the state-of-the-art unsupervised cosegmentation and interactive cosegmentation methods on the iCoseg and MSRC benchmark data sets.

UR - http://www.scopus.com/inward/record.url?scp=84938896314&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84938896314&partnerID=8YFLogxK

U2 - 10.1109/TIP.2015.2456636

DO - 10.1109/TIP.2015.2456636

M3 - Article

AN - SCOPUS:84938896314

VL - 24

SP - 3966

EP - 3977

JO - IEEE Transactions on Image Processing

JF - IEEE Transactions on Image Processing

SN - 1057-7149

IS - 11

M1 - 7156153

ER -