Animals on the web

Tamara L. Berg, David Alexander Forsyth

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

We demonstrate a method for identifying images containing categories of animals. The images we classify depict animals in a wide range of aspects, configurations and appearances. In addition, the images typically portray multiple species that differ in appearance (e.g. ukari's, vervet monkeys, spider monkeys, rhesus monkeys, etc.). Our method is accurate despite this variation and relies on four simple cues: text, color, shape and texture. Visual cues are evaluated by a voting method that compares local image phenomena with a number of visual exemplars for the category. The visual exemplars are obtained using a clustering method applied to text on web pages. The only supervision required involves identifying which clusters of exemplars refer to which sense of a term (for example, "monkey" can refer to an animal or a bandmember). Because our method is applied to web pages with free text, the word cue is extremely noisy. We show unequivocal evidence that visual information improves performance for our task. Our method allows us to produce large, accurate and challenging visual datasets mostly automatically.

Original languageEnglish (US)
Title of host publicationProceedings - 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2006
Pages1463-1470
Number of pages8
DOIs
StatePublished - Dec 22 2006
Event2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2006 - New York, NY, United States
Duration: Jun 17 2006Jun 22 2006

Publication series

NameProceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
Volume2
ISSN (Print)1063-6919

Other

Other2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2006
CountryUnited States
CityNew York, NY
Period6/17/066/22/06

Fingerprint

Animals
Websites
Textures
Color

ASJC Scopus subject areas

  • Software
  • Computer Vision and Pattern Recognition

Cite this

Berg, T. L., & Forsyth, D. A. (2006). Animals on the web. In Proceedings - 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2006 (pp. 1463-1470). [1640929] (Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition; Vol. 2). https://doi.org/10.1109/CVPR.2006.57

Animals on the web. / Berg, Tamara L.; Forsyth, David Alexander.

Proceedings - 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2006. 2006. p. 1463-1470 1640929 (Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition; Vol. 2).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Berg, TL & Forsyth, DA 2006, Animals on the web. in Proceedings - 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2006., 1640929, Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, vol. 2, pp. 1463-1470, 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2006, New York, NY, United States, 6/17/06. https://doi.org/10.1109/CVPR.2006.57
Berg TL, Forsyth DA. Animals on the web. In Proceedings - 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2006. 2006. p. 1463-1470. 1640929. (Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition). https://doi.org/10.1109/CVPR.2006.57
Berg, Tamara L. ; Forsyth, David Alexander. / Animals on the web. Proceedings - 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2006. 2006. pp. 1463-1470 (Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition).
@inproceedings{ee9e4cb19c7f406ca748852825cf7277,
title = "Animals on the web",
abstract = "We demonstrate a method for identifying images containing categories of animals. The images we classify depict animals in a wide range of aspects, configurations and appearances. In addition, the images typically portray multiple species that differ in appearance (e.g. ukari's, vervet monkeys, spider monkeys, rhesus monkeys, etc.). Our method is accurate despite this variation and relies on four simple cues: text, color, shape and texture. Visual cues are evaluated by a voting method that compares local image phenomena with a number of visual exemplars for the category. The visual exemplars are obtained using a clustering method applied to text on web pages. The only supervision required involves identifying which clusters of exemplars refer to which sense of a term (for example, {"}monkey{"} can refer to an animal or a bandmember). Because our method is applied to web pages with free text, the word cue is extremely noisy. We show unequivocal evidence that visual information improves performance for our task. Our method allows us to produce large, accurate and challenging visual datasets mostly automatically.",
author = "Berg, {Tamara L.} and Forsyth, {David Alexander}",
year = "2006",
month = "12",
day = "22",
doi = "10.1109/CVPR.2006.57",
language = "English (US)",
isbn = "0769525970",
series = "Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition",
pages = "1463--1470",
booktitle = "Proceedings - 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2006",

}

TY - GEN

T1 - Animals on the web

AU - Berg, Tamara L.

AU - Forsyth, David Alexander

PY - 2006/12/22

Y1 - 2006/12/22

N2 - We demonstrate a method for identifying images containing categories of animals. The images we classify depict animals in a wide range of aspects, configurations and appearances. In addition, the images typically portray multiple species that differ in appearance (e.g. ukari's, vervet monkeys, spider monkeys, rhesus monkeys, etc.). Our method is accurate despite this variation and relies on four simple cues: text, color, shape and texture. Visual cues are evaluated by a voting method that compares local image phenomena with a number of visual exemplars for the category. The visual exemplars are obtained using a clustering method applied to text on web pages. The only supervision required involves identifying which clusters of exemplars refer to which sense of a term (for example, "monkey" can refer to an animal or a bandmember). Because our method is applied to web pages with free text, the word cue is extremely noisy. We show unequivocal evidence that visual information improves performance for our task. Our method allows us to produce large, accurate and challenging visual datasets mostly automatically.

AB - We demonstrate a method for identifying images containing categories of animals. The images we classify depict animals in a wide range of aspects, configurations and appearances. In addition, the images typically portray multiple species that differ in appearance (e.g. ukari's, vervet monkeys, spider monkeys, rhesus monkeys, etc.). Our method is accurate despite this variation and relies on four simple cues: text, color, shape and texture. Visual cues are evaluated by a voting method that compares local image phenomena with a number of visual exemplars for the category. The visual exemplars are obtained using a clustering method applied to text on web pages. The only supervision required involves identifying which clusters of exemplars refer to which sense of a term (for example, "monkey" can refer to an animal or a bandmember). Because our method is applied to web pages with free text, the word cue is extremely noisy. We show unequivocal evidence that visual information improves performance for our task. Our method allows us to produce large, accurate and challenging visual datasets mostly automatically.

UR - http://www.scopus.com/inward/record.url?scp=33845595951&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33845595951&partnerID=8YFLogxK

U2 - 10.1109/CVPR.2006.57

DO - 10.1109/CVPR.2006.57

M3 - Conference contribution

AN - SCOPUS:33845595951

SN - 0769525970

SN - 9780769525976

T3 - Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition

SP - 1463

EP - 1470

BT - Proceedings - 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2006

ER -