Joint probabilistic techniques for tracking multi-part objects

Christopher Rasmussen, Gregory Hager

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

Abstract

Common objects such as people and cars comprise many visual parts and attributes, yet image-based tracking algorithms are often keyed to only one of a target's identifying characteristics. In this paper, we present a framework for combining and sharing information among several state estimation processes operating on the same underlying visual object. Well-known techniques for joint probabilistic data association are adapted to yield increased robustness when multiple trackers attuned to disparate visual cues are deployed simultaneously. We also formulate a measure of tracker confidence, based on distinctiveness and occlusion probability, which permits the deactivation of trackers before erroneous state estimates adversely affect the ensemble. We will discuss experiments focusing on color-region- and snake-based tracking that demonstrate the efficacy of this approach.

Original languageEnglish (US)
Title of host publicationProceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
PublisherIEEE Comp Soc
Pages16-21
Number of pages6
StatePublished - 1998
Externally publishedYes
EventProceedings of the 1998 IEEE Computer Society Conference on Computer Vision and Pattern Recognition - Santa Barbara, CA, USA
Duration: Jun 23 1998Jun 25 1998

Other

OtherProceedings of the 1998 IEEE Computer Society Conference on Computer Vision and Pattern Recognition
CitySanta Barbara, CA, USA
Period6/23/986/25/98

Fingerprint

State estimation
Railroad cars
Color
Experiments

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition
  • Software
  • Control and Systems Engineering
  • Electrical and Electronic Engineering

Cite this

Rasmussen, C., & Hager, G. (1998). Joint probabilistic techniques for tracking multi-part objects. In Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition (pp. 16-21). IEEE Comp Soc.

Joint probabilistic techniques for tracking multi-part objects. / Rasmussen, Christopher; Hager, Gregory.

Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. IEEE Comp Soc, 1998. p. 16-21.

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

Rasmussen, C & Hager, G 1998, Joint probabilistic techniques for tracking multi-part objects. in Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. IEEE Comp Soc, pp. 16-21, Proceedings of the 1998 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Santa Barbara, CA, USA, 6/23/98.
Rasmussen C, Hager G. Joint probabilistic techniques for tracking multi-part objects. In Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. IEEE Comp Soc. 1998. p. 16-21
Rasmussen, Christopher ; Hager, Gregory. / Joint probabilistic techniques for tracking multi-part objects. Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. IEEE Comp Soc, 1998. pp. 16-21
@inproceedings{735ed5789b7d48fabc9b3cfedef35643,
title = "Joint probabilistic techniques for tracking multi-part objects",
abstract = "Common objects such as people and cars comprise many visual parts and attributes, yet image-based tracking algorithms are often keyed to only one of a target's identifying characteristics. In this paper, we present a framework for combining and sharing information among several state estimation processes operating on the same underlying visual object. Well-known techniques for joint probabilistic data association are adapted to yield increased robustness when multiple trackers attuned to disparate visual cues are deployed simultaneously. We also formulate a measure of tracker confidence, based on distinctiveness and occlusion probability, which permits the deactivation of trackers before erroneous state estimates adversely affect the ensemble. We will discuss experiments focusing on color-region- and snake-based tracking that demonstrate the efficacy of this approach.",
author = "Christopher Rasmussen and Gregory Hager",
year = "1998",
language = "English (US)",
pages = "16--21",
booktitle = "Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition",
publisher = "IEEE Comp Soc",

}

TY - GEN

T1 - Joint probabilistic techniques for tracking multi-part objects

AU - Rasmussen, Christopher

AU - Hager, Gregory

PY - 1998

Y1 - 1998

N2 - Common objects such as people and cars comprise many visual parts and attributes, yet image-based tracking algorithms are often keyed to only one of a target's identifying characteristics. In this paper, we present a framework for combining and sharing information among several state estimation processes operating on the same underlying visual object. Well-known techniques for joint probabilistic data association are adapted to yield increased robustness when multiple trackers attuned to disparate visual cues are deployed simultaneously. We also formulate a measure of tracker confidence, based on distinctiveness and occlusion probability, which permits the deactivation of trackers before erroneous state estimates adversely affect the ensemble. We will discuss experiments focusing on color-region- and snake-based tracking that demonstrate the efficacy of this approach.

AB - Common objects such as people and cars comprise many visual parts and attributes, yet image-based tracking algorithms are often keyed to only one of a target's identifying characteristics. In this paper, we present a framework for combining and sharing information among several state estimation processes operating on the same underlying visual object. Well-known techniques for joint probabilistic data association are adapted to yield increased robustness when multiple trackers attuned to disparate visual cues are deployed simultaneously. We also formulate a measure of tracker confidence, based on distinctiveness and occlusion probability, which permits the deactivation of trackers before erroneous state estimates adversely affect the ensemble. We will discuss experiments focusing on color-region- and snake-based tracking that demonstrate the efficacy of this approach.

UR - http://www.scopus.com/inward/record.url?scp=0032299329&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0032299329&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0032299329

SP - 16

EP - 21

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

PB - IEEE Comp Soc

ER -