Finding pictures of objects in large collections of images

David A. Forsyth, Jitendra Malik, Margaret M. Fleck, Hayit Greenspan, Thomas Leung, Serge Belongie, Chad Carson, Chris Bregler

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

Abstract

Retrieving images from very large collections, using image content as a key, is becoming an important problem. Users prefer to ask for pictures using notions of content that are strongly oriented to the presence of abstractly defined objects. Computer programs that implement these queries automatically are desirable, but are hard to build because conventional object recognition techniques from computer vision cannot recognize very general objects in very general contexts. This paper describes our approach to object recognition, which is structured around a sequence of increasingly specialized grouping activities that assemble coherent regions of image that can be shown to satisfy increasingly stringent constraints. The constraints that are satisfied provide a form of object classification in quite general contexts. This view of recognition is distinguished by: far richer involvement of early visual primitives, including color and texture; hierarchical grouping and learning strategies in the classification process; the ability to deal with rather general objects in uncontrolled configurations and contexts. We illustrate these properties with four case-studies: one demonstrating the use of color and texture descriptors; one showing how trees can be described by fusing texture and geometric properties; one learning scenery concepts using grouped features; and one showing how this view of recognition yields a program that can tell, quite accurately, whether a picture contains naked people or not.

Original languageEnglish (US)
Title of host publicationObject Representation in Computer Vision II - ECCV 1996 International Workshop, Proceedings
EditorsAndrew Zisserman, Jean Ponce, Martial Hebert, Martial Hebert, Martial Hebert, Jean Ponce, Andrew Zisserman, Jean Ponce, Andrew Zisserman
PublisherSpringer-Verlag
Pages335-360
Number of pages26
ISBN (Print)3540617507, 3540617507, 3540617507, 9783540617501, 9783540617501, 9783540617501
StatePublished - Jan 1 1996
EventInternational Workshop on Object Representation in Computer Vision II, ECCV 1996 - Cambridge, United Kingdom
Duration: Apr 13 1996Apr 14 1996

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume1144
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

OtherInternational Workshop on Object Representation in Computer Vision II, ECCV 1996
CountryUnited Kingdom
CityCambridge
Period4/13/964/14/96

Fingerprint

Texture
Textures
Object recognition
Object Recognition
Grouping
Concept Learning
Color
Object Classification
Learning Strategies
Computer Vision
Computer vision
Descriptors
Computer program listings
Query
Configuration
Object
Context
Form
Vision

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Forsyth, D. A., Malik, J., Fleck, M. M., Greenspan, H., Leung, T., Belongie, S., ... Bregler, C. (1996). Finding pictures of objects in large collections of images. In A. Zisserman, J. Ponce, M. Hebert, M. Hebert, M. Hebert, J. Ponce, A. Zisserman, J. Ponce, ... A. Zisserman (Eds.), Object Representation in Computer Vision II - ECCV 1996 International Workshop, Proceedings (pp. 335-360). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 1144). Springer-Verlag.

Finding pictures of objects in large collections of images. / Forsyth, David A.; Malik, Jitendra; Fleck, Margaret M.; Greenspan, Hayit; Leung, Thomas; Belongie, Serge; Carson, Chad; Bregler, Chris.

Object Representation in Computer Vision II - ECCV 1996 International Workshop, Proceedings. ed. / Andrew Zisserman; Jean Ponce; Martial Hebert; Martial Hebert; Martial Hebert; Jean Ponce; Andrew Zisserman; Jean Ponce; Andrew Zisserman. Springer-Verlag, 1996. p. 335-360 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 1144).

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

Forsyth, DA, Malik, J, Fleck, MM, Greenspan, H, Leung, T, Belongie, S, Carson, C & Bregler, C 1996, Finding pictures of objects in large collections of images. in A Zisserman, J Ponce, M Hebert, M Hebert, M Hebert, J Ponce, A Zisserman, J Ponce & A Zisserman (eds), Object Representation in Computer Vision II - ECCV 1996 International Workshop, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 1144, Springer-Verlag, pp. 335-360, International Workshop on Object Representation in Computer Vision II, ECCV 1996, Cambridge, United Kingdom, 4/13/96.
Forsyth DA, Malik J, Fleck MM, Greenspan H, Leung T, Belongie S et al. Finding pictures of objects in large collections of images. In Zisserman A, Ponce J, Hebert M, Hebert M, Hebert M, Ponce J, Zisserman A, Ponce J, Zisserman A, editors, Object Representation in Computer Vision II - ECCV 1996 International Workshop, Proceedings. Springer-Verlag. 1996. p. 335-360. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
Forsyth, David A. ; Malik, Jitendra ; Fleck, Margaret M. ; Greenspan, Hayit ; Leung, Thomas ; Belongie, Serge ; Carson, Chad ; Bregler, Chris. / Finding pictures of objects in large collections of images. Object Representation in Computer Vision II - ECCV 1996 International Workshop, Proceedings. editor / Andrew Zisserman ; Jean Ponce ; Martial Hebert ; Martial Hebert ; Martial Hebert ; Jean Ponce ; Andrew Zisserman ; Jean Ponce ; Andrew Zisserman. Springer-Verlag, 1996. pp. 335-360 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{8da13ea815c544aaad9070bd6e0733ad,
title = "Finding pictures of objects in large collections of images",
abstract = "Retrieving images from very large collections, using image content as a key, is becoming an important problem. Users prefer to ask for pictures using notions of content that are strongly oriented to the presence of abstractly defined objects. Computer programs that implement these queries automatically are desirable, but are hard to build because conventional object recognition techniques from computer vision cannot recognize very general objects in very general contexts. This paper describes our approach to object recognition, which is structured around a sequence of increasingly specialized grouping activities that assemble coherent regions of image that can be shown to satisfy increasingly stringent constraints. The constraints that are satisfied provide a form of object classification in quite general contexts. This view of recognition is distinguished by: far richer involvement of early visual primitives, including color and texture; hierarchical grouping and learning strategies in the classification process; the ability to deal with rather general objects in uncontrolled configurations and contexts. We illustrate these properties with four case-studies: one demonstrating the use of color and texture descriptors; one showing how trees can be described by fusing texture and geometric properties; one learning scenery concepts using grouped features; and one showing how this view of recognition yields a program that can tell, quite accurately, whether a picture contains naked people or not.",
author = "Forsyth, {David A.} and Jitendra Malik and Fleck, {Margaret M.} and Hayit Greenspan and Thomas Leung and Serge Belongie and Chad Carson and Chris Bregler",
year = "1996",
month = "1",
day = "1",
language = "English (US)",
isbn = "3540617507",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer-Verlag",
pages = "335--360",
editor = "Andrew Zisserman and Jean Ponce and Martial Hebert and Martial Hebert and Martial Hebert and Jean Ponce and Andrew Zisserman and Jean Ponce and Andrew Zisserman",
booktitle = "Object Representation in Computer Vision II - ECCV 1996 International Workshop, Proceedings",

}

TY - GEN

T1 - Finding pictures of objects in large collections of images

AU - Forsyth, David A.

AU - Malik, Jitendra

AU - Fleck, Margaret M.

AU - Greenspan, Hayit

AU - Leung, Thomas

AU - Belongie, Serge

AU - Carson, Chad

AU - Bregler, Chris

PY - 1996/1/1

Y1 - 1996/1/1

N2 - Retrieving images from very large collections, using image content as a key, is becoming an important problem. Users prefer to ask for pictures using notions of content that are strongly oriented to the presence of abstractly defined objects. Computer programs that implement these queries automatically are desirable, but are hard to build because conventional object recognition techniques from computer vision cannot recognize very general objects in very general contexts. This paper describes our approach to object recognition, which is structured around a sequence of increasingly specialized grouping activities that assemble coherent regions of image that can be shown to satisfy increasingly stringent constraints. The constraints that are satisfied provide a form of object classification in quite general contexts. This view of recognition is distinguished by: far richer involvement of early visual primitives, including color and texture; hierarchical grouping and learning strategies in the classification process; the ability to deal with rather general objects in uncontrolled configurations and contexts. We illustrate these properties with four case-studies: one demonstrating the use of color and texture descriptors; one showing how trees can be described by fusing texture and geometric properties; one learning scenery concepts using grouped features; and one showing how this view of recognition yields a program that can tell, quite accurately, whether a picture contains naked people or not.

AB - Retrieving images from very large collections, using image content as a key, is becoming an important problem. Users prefer to ask for pictures using notions of content that are strongly oriented to the presence of abstractly defined objects. Computer programs that implement these queries automatically are desirable, but are hard to build because conventional object recognition techniques from computer vision cannot recognize very general objects in very general contexts. This paper describes our approach to object recognition, which is structured around a sequence of increasingly specialized grouping activities that assemble coherent regions of image that can be shown to satisfy increasingly stringent constraints. The constraints that are satisfied provide a form of object classification in quite general contexts. This view of recognition is distinguished by: far richer involvement of early visual primitives, including color and texture; hierarchical grouping and learning strategies in the classification process; the ability to deal with rather general objects in uncontrolled configurations and contexts. We illustrate these properties with four case-studies: one demonstrating the use of color and texture descriptors; one showing how trees can be described by fusing texture and geometric properties; one learning scenery concepts using grouped features; and one showing how this view of recognition yields a program that can tell, quite accurately, whether a picture contains naked people or not.

UR - http://www.scopus.com/inward/record.url?scp=84979074103&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84979074103&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84979074103

SN - 3540617507

SN - 3540617507

SN - 3540617507

SN - 9783540617501

SN - 9783540617501

SN - 9783540617501

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 335

EP - 360

BT - Object Representation in Computer Vision II - ECCV 1996 International Workshop, Proceedings

A2 - Zisserman, Andrew

A2 - Ponce, Jean

A2 - Hebert, Martial

A2 - Hebert, Martial

A2 - Hebert, Martial

A2 - Ponce, Jean

A2 - Zisserman, Andrew

A2 - Ponce, Jean

A2 - Zisserman, Andrew

PB - Springer-Verlag

ER -