Hierarchical conditional random field for multi-class image classification

Michael Ying Yang; Wolfgang Förstner; Martin Drauschke

doi:10.5220/0002877404640469

Hierarchical conditional random field for multi-class image classification

Michael Ying Yang, Wolfgang Förstner, Martin Drauschke

Department of Earth Observation Science

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

10 Citations (Scopus)

8 Downloads (Pure)

Abstract

Multi-class image classification has made significant advances in recent years through the combination of local and global features. This paper proposes a novel approach called hierarchical conditional random field (HCRF) that explicitly models region adjacency graph and region hierarchy graph structure of an image. This allows to set up a joint and hierarchical model of local and global discriminative methods that augments conditional random field to a multi-layer model. Region hierarchy graph is based on a multi-scale watershed segmentation.

Original language	English
Title of host publication	VISAPP 2010 - Proceedings of the International Conference on Computer Vision Theory and Applications
Pages	464-469
Number of pages	6
Volume	2
DOIs	https://doi.org/10.5220/0002877404640469
Publication status	Published - 10 Sept 2010
Event	5th International Conference on Computer Vision Theory and Applications, VISAPP 2010 - Angers, France Duration: 17 May 2010 → 21 May 2010 Conference number: 5

Conference

Conference	5th International Conference on Computer Vision Theory and Applications, VISAPP 2010
Abbreviated title	VISAPP 2010
Country/Territory	France
City	Angers
Period	17/05/10 → 21/05/10

Keywords

Hierarchical conditional random field
Image segmentation
Multi-class image classification
Region adjacency graph
Region hierarchy graph

Access to Document

10.5220/0002877404640469Licence: CC BY-NC-ND

ArticleFinal published version, 1.23 MBLicence: CC BY-NC-ND

Cite this

@inproceedings{32c0af5dd85843f69644d0109fdc770d,

title = "Hierarchical conditional random field for multi-class image classification",

abstract = "Multi-class image classification has made significant advances in recent years through the combination of local and global features. This paper proposes a novel approach called hierarchical conditional random field (HCRF) that explicitly models region adjacency graph and region hierarchy graph structure of an image. This allows to set up a joint and hierarchical model of local and global discriminative methods that augments conditional random field to a multi-layer model. Region hierarchy graph is based on a multi-scale watershed segmentation.",

keywords = "Hierarchical conditional random field, Image segmentation, Multi-class image classification, Region adjacency graph, Region hierarchy graph",

author = "Yang, {Michael Ying} and Wolfgang F{\"o}rstner and Martin Drauschke",

year = "2010",

month = sep,

day = "10",

doi = "10.5220/0002877404640469",

language = "English",

isbn = "9789896740283",

volume = "2",

pages = "464--469",

booktitle = "VISAPP 2010 - Proceedings of the International Conference on Computer Vision Theory and Applications",

note = "5th International Conference on Computer Vision Theory and Applications, VISAPP 2010, VISAPP 2010 ; Conference date: 17-05-2010 Through 21-05-2010",

}

Yang, MY, Förstner, W & Drauschke, M 2010, Hierarchical conditional random field for multi-class image classification. in VISAPP 2010 - Proceedings of the International Conference on Computer Vision Theory and Applications. vol. 2, pp. 464-469, 5th International Conference on Computer Vision Theory and Applications, VISAPP 2010, Angers, France, 17/05/10. https://doi.org/10.5220/0002877404640469

Hierarchical conditional random field for multi-class image classification. / Yang, Michael Ying; Förstner, Wolfgang; Drauschke, Martin.
VISAPP 2010 - Proceedings of the International Conference on Computer Vision Theory and Applications. Vol. 2 2010. p. 464-469.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Hierarchical conditional random field for multi-class image classification

AU - Yang, Michael Ying

AU - Förstner, Wolfgang

AU - Drauschke, Martin

N1 - Conference code: 5

PY - 2010/9/10

Y1 - 2010/9/10

N2 - Multi-class image classification has made significant advances in recent years through the combination of local and global features. This paper proposes a novel approach called hierarchical conditional random field (HCRF) that explicitly models region adjacency graph and region hierarchy graph structure of an image. This allows to set up a joint and hierarchical model of local and global discriminative methods that augments conditional random field to a multi-layer model. Region hierarchy graph is based on a multi-scale watershed segmentation.

AB - Multi-class image classification has made significant advances in recent years through the combination of local and global features. This paper proposes a novel approach called hierarchical conditional random field (HCRF) that explicitly models region adjacency graph and region hierarchy graph structure of an image. This allows to set up a joint and hierarchical model of local and global discriminative methods that augments conditional random field to a multi-layer model. Region hierarchy graph is based on a multi-scale watershed segmentation.

KW - Hierarchical conditional random field

KW - Image segmentation

KW - Multi-class image classification

KW - Region adjacency graph

KW - Region hierarchy graph

UR - http://www.scopus.com/inward/record.url?scp=77956324307&partnerID=8YFLogxK

U2 - 10.5220/0002877404640469

DO - 10.5220/0002877404640469

M3 - Conference contribution

AN - SCOPUS:77956324307

SN - 9789896740283

VL - 2

SP - 464

EP - 469

BT - VISAPP 2010 - Proceedings of the International Conference on Computer Vision Theory and Applications

T2 - 5th International Conference on Computer Vision Theory and Applications, VISAPP 2010

Y2 - 17 May 2010 through 21 May 2010

ER -

Hierarchical conditional random field for multi-class image classification

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this