Constructing Visual Taxonomies by Shape

Anthony Cook; M. J. Gibbens

doi:10.1109/ICPR.2006.405

Constructing Visual Taxonomies by Shape

Anthony Cook, M. J. Gibbens

Department of Physics

Research output: Chapter in Book/Report/Conference proceeding › Conference Proceeding (Non-Journal item)

1 Citation (Scopus)

Abstract

We investigate the use of statistical shape measures for segmented image regions to construct taxonomies of visual similarity. It is demonstrated that without the use of a priori knowledge, cluster analysis can be used to impose structure on heterogeneous image data sets. We develop visual taxonomies to accomplish moderate classification tasks, and provide a framework for more powerful, open-ended analysis of large data sets. The power of this method is demonstrated using a visual taxonomy of textual data, which is shown to be efficient in an MDL context

Original language	English
Title of host publication	18th International Conference on Pattern Recognition, 2006. ICPR 2006
Publisher	IEEE Press
Pages	732 - 735
Number of pages	4
ISBN (Print)	0-7695-2521-0
DOIs	https://doi.org/10.1109/ICPR.2006.405
Publication status	Published - 2006

Access to Document

10.1109/ICPR.2006.405

http://hdl.handle.net/2160/1507

Permanent link

Persistent link

Cite this

@inproceedings{21631f51cf654bbc80a7cbf129d1c910,

title = "Constructing Visual Taxonomies by Shape",

abstract = "We investigate the use of statistical shape measures for segmented image regions to construct taxonomies of visual similarity. It is demonstrated that without the use of a priori knowledge, cluster analysis can be used to impose structure on heterogeneous image data sets. We develop visual taxonomies to accomplish moderate classification tasks, and provide a framework for more powerful, open-ended analysis of large data sets. The power of this method is demonstrated using a visual taxonomy of textual data, which is shown to be efficient in an MDL context",

author = "Anthony Cook and Gibbens, {M. J.}",

note = "Cook, Anthony; Gibbens, M.J., (2006) 'Constructing Visual Taxonomies by Shape', 18th International Conference on Pattern Recognition (ICPR'06) Volume 2, pp. 732 - 735 RAE2008",

year = "2006",

doi = "10.1109/ICPR.2006.405",

language = "English",

isbn = "0-7695-2521-0",

pages = "732 -- 735",

booktitle = "18th International Conference on Pattern Recognition, 2006. ICPR 2006",

publisher = "IEEE Press",

address = "United States of America",

}

TY - GEN

T1 - Constructing Visual Taxonomies by Shape

AU - Cook, Anthony

AU - Gibbens, M. J.

N1 - Cook, Anthony; Gibbens, M.J., (2006) 'Constructing Visual Taxonomies by Shape', 18th International Conference on Pattern Recognition (ICPR'06) Volume 2, pp. 732 - 735 RAE2008

PY - 2006

Y1 - 2006

N2 - We investigate the use of statistical shape measures for segmented image regions to construct taxonomies of visual similarity. It is demonstrated that without the use of a priori knowledge, cluster analysis can be used to impose structure on heterogeneous image data sets. We develop visual taxonomies to accomplish moderate classification tasks, and provide a framework for more powerful, open-ended analysis of large data sets. The power of this method is demonstrated using a visual taxonomy of textual data, which is shown to be efficient in an MDL context

AB - We investigate the use of statistical shape measures for segmented image regions to construct taxonomies of visual similarity. It is demonstrated that without the use of a priori knowledge, cluster analysis can be used to impose structure on heterogeneous image data sets. We develop visual taxonomies to accomplish moderate classification tasks, and provide a framework for more powerful, open-ended analysis of large data sets. The power of this method is demonstrated using a visual taxonomy of textual data, which is shown to be efficient in an MDL context

U2 - 10.1109/ICPR.2006.405

DO - 10.1109/ICPR.2006.405

M3 - Conference Proceeding (Non-Journal item)

SN - 0-7695-2521-0

SP - 732

EP - 735

BT - 18th International Conference on Pattern Recognition, 2006. ICPR 2006

PB - IEEE Press

ER -

Constructing Visual Taxonomies by Shape

Abstract

Access to Document

Permanent link

Fingerprint

Cite this