Beyond pixels: a comprehensive survey from bottom-up to semantic image segmentation and cosegmentation

Hongyuan Zhu, Fanman Meng, Jianfei Cai, Shijian Lu

Research output: Contribution to journalArticleResearchpeer-review

93 Citations (Scopus)

Abstract

Image segmentation refers to the process to divide an image into meaningful non-overlapping regions according to human perception, which has become a classic topic since the early ages of computer vision. A lot of research has been conducted and has resulted in many applications. While many segmentation algorithms exist, there are only a few sparse and outdated summarizations available. Thus, in this paper, we aim to provide a comprehensive review of the recent progress in the field. Covering 190 publications, we give an overview of broad segmentation topics including not only the classic unsupervised methods, but also the recent weakly-/semi-supervised methods and the fully-supervised methods. In addition, we review the existing influential datasets and evaluation metrics. We also suggest some design choices and research directions for future research in image segmentation.

Original languageEnglish
Pages (from-to)12-27
Number of pages16
JournalJournal of Visual Communication and Image Representation
Volume34
DOIs
Publication statusPublished - Jan 2016
Externally publishedYes

Keywords

  • Image cosegmentation
  • Image segmentation
  • Interactive image segmentation
  • Object proposal
  • Semantic image parsing
  • Superpixel
  • Unsupervised image segmentation
  • Weakly-supervised image segmentation

Cite this

@article{6cdc73080974400cbfbfccc8a544a6c6,
title = "Beyond pixels: a comprehensive survey from bottom-up to semantic image segmentation and cosegmentation",
abstract = "Image segmentation refers to the process to divide an image into meaningful non-overlapping regions according to human perception, which has become a classic topic since the early ages of computer vision. A lot of research has been conducted and has resulted in many applications. While many segmentation algorithms exist, there are only a few sparse and outdated summarizations available. Thus, in this paper, we aim to provide a comprehensive review of the recent progress in the field. Covering 190 publications, we give an overview of broad segmentation topics including not only the classic unsupervised methods, but also the recent weakly-/semi-supervised methods and the fully-supervised methods. In addition, we review the existing influential datasets and evaluation metrics. We also suggest some design choices and research directions for future research in image segmentation.",
keywords = "Image cosegmentation, Image segmentation, Interactive image segmentation, Object proposal, Semantic image parsing, Superpixel, Unsupervised image segmentation, Weakly-supervised image segmentation",
author = "Hongyuan Zhu and Fanman Meng and Jianfei Cai and Shijian Lu",
year = "2016",
month = "1",
doi = "10.1016/j.jvcir.2015.10.012",
language = "English",
volume = "34",
pages = "12--27",
journal = "Journal of Visual Communication and Image Representation",
issn = "1047-3203",
publisher = "Elsevier",

}

Beyond pixels : a comprehensive survey from bottom-up to semantic image segmentation and cosegmentation. / Zhu, Hongyuan; Meng, Fanman; Cai, Jianfei; Lu, Shijian.

In: Journal of Visual Communication and Image Representation, Vol. 34, 01.2016, p. 12-27.

Research output: Contribution to journalArticleResearchpeer-review

TY - JOUR

T1 - Beyond pixels

T2 - a comprehensive survey from bottom-up to semantic image segmentation and cosegmentation

AU - Zhu, Hongyuan

AU - Meng, Fanman

AU - Cai, Jianfei

AU - Lu, Shijian

PY - 2016/1

Y1 - 2016/1

N2 - Image segmentation refers to the process to divide an image into meaningful non-overlapping regions according to human perception, which has become a classic topic since the early ages of computer vision. A lot of research has been conducted and has resulted in many applications. While many segmentation algorithms exist, there are only a few sparse and outdated summarizations available. Thus, in this paper, we aim to provide a comprehensive review of the recent progress in the field. Covering 190 publications, we give an overview of broad segmentation topics including not only the classic unsupervised methods, but also the recent weakly-/semi-supervised methods and the fully-supervised methods. In addition, we review the existing influential datasets and evaluation metrics. We also suggest some design choices and research directions for future research in image segmentation.

AB - Image segmentation refers to the process to divide an image into meaningful non-overlapping regions according to human perception, which has become a classic topic since the early ages of computer vision. A lot of research has been conducted and has resulted in many applications. While many segmentation algorithms exist, there are only a few sparse and outdated summarizations available. Thus, in this paper, we aim to provide a comprehensive review of the recent progress in the field. Covering 190 publications, we give an overview of broad segmentation topics including not only the classic unsupervised methods, but also the recent weakly-/semi-supervised methods and the fully-supervised methods. In addition, we review the existing influential datasets and evaluation metrics. We also suggest some design choices and research directions for future research in image segmentation.

KW - Image cosegmentation

KW - Image segmentation

KW - Interactive image segmentation

KW - Object proposal

KW - Semantic image parsing

KW - Superpixel

KW - Unsupervised image segmentation

KW - Weakly-supervised image segmentation

UR - http://www.scopus.com/inward/record.url?scp=84946771932&partnerID=8YFLogxK

U2 - 10.1016/j.jvcir.2015.10.012

DO - 10.1016/j.jvcir.2015.10.012

M3 - Article

AN - SCOPUS:84946771932

VL - 34

SP - 12

EP - 27

JO - Journal of Visual Communication and Image Representation

JF - Journal of Visual Communication and Image Representation

SN - 1047-3203

ER -