Finding people by sampling

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

Abstract

We show how to use a sampling method to find sparsely clad people in static images. People are modeled as an assembly of nine cylindrical segments. Segments are found using an EM algorithm, and then assembled into hypotheses incrementally, using a learned likelihood model. Each assembly step passes on a set of samples of its likelihood to the next; this yields effective pruning of the space of hypotheses. The collection of available nine-segment hypotheses is then represented by a set of equivalence classes, which yield an efficient pruning process. The posterior for the number of people is obtained from the class representatives. People are counted quite accurately in images of real scenes using an MAP estimate. We show the method allows top-down as well as bottom up reasoning. While the method can be overwhelmed by very large numbers of segments, we show that this problem can be avoided by quite simple pruning steps.

Original languageEnglish (US)
Title of host publicationProceedings of the IEEE International Conference on Computer Vision
PublisherIEEE
Pages1092-1097
Number of pages6
Volume2
StatePublished - 1999
Externally publishedYes
EventProceedings of the 1999 7th IEEE International Conference on Computer Vision (ICCV'99) - Kerkyra, Greece
Duration: Sep 20 1999Sep 27 1999

Other

OtherProceedings of the 1999 7th IEEE International Conference on Computer Vision (ICCV'99)
CityKerkyra, Greece
Period9/20/999/27/99

Fingerprint

Sampling
Equivalence classes

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition
  • Software

Cite this

Ioffe, S., & Forsyth, D. A. (1999). Finding people by sampling. In Proceedings of the IEEE International Conference on Computer Vision (Vol. 2, pp. 1092-1097). IEEE.

Finding people by sampling. / Ioffe, Sergey; Forsyth, David Alexander.

Proceedings of the IEEE International Conference on Computer Vision. Vol. 2 IEEE, 1999. p. 1092-1097.

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

Ioffe, S & Forsyth, DA 1999, Finding people by sampling. in Proceedings of the IEEE International Conference on Computer Vision. vol. 2, IEEE, pp. 1092-1097, Proceedings of the 1999 7th IEEE International Conference on Computer Vision (ICCV'99), Kerkyra, Greece, 9/20/99.
Ioffe S, Forsyth DA. Finding people by sampling. In Proceedings of the IEEE International Conference on Computer Vision. Vol. 2. IEEE. 1999. p. 1092-1097
Ioffe, Sergey ; Forsyth, David Alexander. / Finding people by sampling. Proceedings of the IEEE International Conference on Computer Vision. Vol. 2 IEEE, 1999. pp. 1092-1097
@inproceedings{04ebd4fc1dda48cbb6b09fa1ecb1066c,
title = "Finding people by sampling",
abstract = "We show how to use a sampling method to find sparsely clad people in static images. People are modeled as an assembly of nine cylindrical segments. Segments are found using an EM algorithm, and then assembled into hypotheses incrementally, using a learned likelihood model. Each assembly step passes on a set of samples of its likelihood to the next; this yields effective pruning of the space of hypotheses. The collection of available nine-segment hypotheses is then represented by a set of equivalence classes, which yield an efficient pruning process. The posterior for the number of people is obtained from the class representatives. People are counted quite accurately in images of real scenes using an MAP estimate. We show the method allows top-down as well as bottom up reasoning. While the method can be overwhelmed by very large numbers of segments, we show that this problem can be avoided by quite simple pruning steps.",
author = "Sergey Ioffe and Forsyth, {David Alexander}",
year = "1999",
language = "English (US)",
volume = "2",
pages = "1092--1097",
booktitle = "Proceedings of the IEEE International Conference on Computer Vision",
publisher = "IEEE",

}

TY - GEN

T1 - Finding people by sampling

AU - Ioffe, Sergey

AU - Forsyth, David Alexander

PY - 1999

Y1 - 1999

N2 - We show how to use a sampling method to find sparsely clad people in static images. People are modeled as an assembly of nine cylindrical segments. Segments are found using an EM algorithm, and then assembled into hypotheses incrementally, using a learned likelihood model. Each assembly step passes on a set of samples of its likelihood to the next; this yields effective pruning of the space of hypotheses. The collection of available nine-segment hypotheses is then represented by a set of equivalence classes, which yield an efficient pruning process. The posterior for the number of people is obtained from the class representatives. People are counted quite accurately in images of real scenes using an MAP estimate. We show the method allows top-down as well as bottom up reasoning. While the method can be overwhelmed by very large numbers of segments, we show that this problem can be avoided by quite simple pruning steps.

AB - We show how to use a sampling method to find sparsely clad people in static images. People are modeled as an assembly of nine cylindrical segments. Segments are found using an EM algorithm, and then assembled into hypotheses incrementally, using a learned likelihood model. Each assembly step passes on a set of samples of its likelihood to the next; this yields effective pruning of the space of hypotheses. The collection of available nine-segment hypotheses is then represented by a set of equivalence classes, which yield an efficient pruning process. The posterior for the number of people is obtained from the class representatives. People are counted quite accurately in images of real scenes using an MAP estimate. We show the method allows top-down as well as bottom up reasoning. While the method can be overwhelmed by very large numbers of segments, we show that this problem can be avoided by quite simple pruning steps.

UR - http://www.scopus.com/inward/record.url?scp=0033284441&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0033284441&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0033284441

VL - 2

SP - 1092

EP - 1097

BT - Proceedings of the IEEE International Conference on Computer Vision

PB - IEEE

ER -