Learning to find pictures of people

Sergey Ioffe, David Forsyth

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

Abstract

Finding articulated objects, like people, in pictures presents a particularly difficult object recognition problem. We show how to find people by finding putative body segments, and then constructing assemblies of those segments that are consistent with the constraints on the appearance of a person that result from kinematic properties. Since a reasonable model of a person requires at least nine segments, it is not possible to present every group to a classifier. Instead, the search can be pruned by using projected versions of a classifier that accepts groups corresponding to people. We describe an efficient projection algorithm for one popular classifier, and demonstrate that our approach can be used to determine whether images of real scenes contain people.

Original languageEnglish (US)
Title of host publicationAdvances in Neural Information Processing Systems 11 - Proceedings of the 1998 Conference, NIPS 1998
PublisherNeural information processing systems foundation
Pages782-788
Number of pages7
ISBN (Print)0262112450, 9780262112451
StatePublished - Jan 1 1999
Externally publishedYes
Event12th Annual Conference on Neural Information Processing Systems, NIPS 1998 - Denver, CO, United States
Duration: Nov 30 1998Dec 5 1998

Publication series

NameAdvances in Neural Information Processing Systems
ISSN (Print)1049-5258

Other

Other12th Annual Conference on Neural Information Processing Systems, NIPS 1998
CountryUnited States
CityDenver, CO
Period11/30/9812/5/98

Fingerprint

Classifiers
Object recognition
Kinematics

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Information Systems
  • Signal Processing

Cite this

Ioffe, S., & Forsyth, D. (1999). Learning to find pictures of people. In Advances in Neural Information Processing Systems 11 - Proceedings of the 1998 Conference, NIPS 1998 (pp. 782-788). (Advances in Neural Information Processing Systems). Neural information processing systems foundation.

Learning to find pictures of people. / Ioffe, Sergey; Forsyth, David.

Advances in Neural Information Processing Systems 11 - Proceedings of the 1998 Conference, NIPS 1998. Neural information processing systems foundation, 1999. p. 782-788 (Advances in Neural Information Processing Systems).

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

Ioffe, S & Forsyth, D 1999, Learning to find pictures of people. in Advances in Neural Information Processing Systems 11 - Proceedings of the 1998 Conference, NIPS 1998. Advances in Neural Information Processing Systems, Neural information processing systems foundation, pp. 782-788, 12th Annual Conference on Neural Information Processing Systems, NIPS 1998, Denver, CO, United States, 11/30/98.
Ioffe S, Forsyth D. Learning to find pictures of people. In Advances in Neural Information Processing Systems 11 - Proceedings of the 1998 Conference, NIPS 1998. Neural information processing systems foundation. 1999. p. 782-788. (Advances in Neural Information Processing Systems).
Ioffe, Sergey ; Forsyth, David. / Learning to find pictures of people. Advances in Neural Information Processing Systems 11 - Proceedings of the 1998 Conference, NIPS 1998. Neural information processing systems foundation, 1999. pp. 782-788 (Advances in Neural Information Processing Systems).
@inproceedings{f8bccaf9ed724771ad614c4d860eaa56,
title = "Learning to find pictures of people",
abstract = "Finding articulated objects, like people, in pictures presents a particularly difficult object recognition problem. We show how to find people by finding putative body segments, and then constructing assemblies of those segments that are consistent with the constraints on the appearance of a person that result from kinematic properties. Since a reasonable model of a person requires at least nine segments, it is not possible to present every group to a classifier. Instead, the search can be pruned by using projected versions of a classifier that accepts groups corresponding to people. We describe an efficient projection algorithm for one popular classifier, and demonstrate that our approach can be used to determine whether images of real scenes contain people.",
author = "Sergey Ioffe and David Forsyth",
year = "1999",
month = "1",
day = "1",
language = "English (US)",
isbn = "0262112450",
series = "Advances in Neural Information Processing Systems",
publisher = "Neural information processing systems foundation",
pages = "782--788",
booktitle = "Advances in Neural Information Processing Systems 11 - Proceedings of the 1998 Conference, NIPS 1998",

}

TY - GEN

T1 - Learning to find pictures of people

AU - Ioffe, Sergey

AU - Forsyth, David

PY - 1999/1/1

Y1 - 1999/1/1

N2 - Finding articulated objects, like people, in pictures presents a particularly difficult object recognition problem. We show how to find people by finding putative body segments, and then constructing assemblies of those segments that are consistent with the constraints on the appearance of a person that result from kinematic properties. Since a reasonable model of a person requires at least nine segments, it is not possible to present every group to a classifier. Instead, the search can be pruned by using projected versions of a classifier that accepts groups corresponding to people. We describe an efficient projection algorithm for one popular classifier, and demonstrate that our approach can be used to determine whether images of real scenes contain people.

AB - Finding articulated objects, like people, in pictures presents a particularly difficult object recognition problem. We show how to find people by finding putative body segments, and then constructing assemblies of those segments that are consistent with the constraints on the appearance of a person that result from kinematic properties. Since a reasonable model of a person requires at least nine segments, it is not possible to present every group to a classifier. Instead, the search can be pruned by using projected versions of a classifier that accepts groups corresponding to people. We describe an efficient projection algorithm for one popular classifier, and demonstrate that our approach can be used to determine whether images of real scenes contain people.

UR - http://www.scopus.com/inward/record.url?scp=34247628197&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=34247628197&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:34247628197

SN - 0262112450

SN - 9780262112451

T3 - Advances in Neural Information Processing Systems

SP - 782

EP - 788

BT - Advances in Neural Information Processing Systems 11 - Proceedings of the 1998 Conference, NIPS 1998

PB - Neural information processing systems foundation

ER -