Adaptive learning algorithm for SVM applied to feature tracking

Ashutosh Garg, Ira Cohen, Thomas S Huang

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

Abstract

The framework of support vector machines (SVM) is becoming extremely popular in the field of statistical pattern classification. In this paper we investigate a technique which couples Kalman filter closely with the SVM. The problem of object tracking can be seen as a pattern recognition problem. However, because of the dynamics, this pattern might experience some changes over time. In order to keep track of the position of the pattern and to make out the desired pattern from the background, we must have some strong continuous time model. We propose an algorithm which combines the Markov property of the Kalman filter with the strong classification capability of SVM. The whole system has been tested on real life problems and we found that with this framework we could track a particular object even in a frame which contains identical objects. The results were compared to that of obtained by color blob tracking which showed the strength of the approach.

Original languageEnglish (US)
Title of host publicationProceedings - 1999 International Conference on Information Intelligence and Systems, ICIIS 1999
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages388-395
Number of pages8
ISBN (Electronic)0769504469, 9780769504469
DOIs
StatePublished - Jan 1 1999
Event1999 International Conference on Information Intelligence and Systems, ICIIS 1999 - Bethesda, United States
Duration: Oct 31 1999Nov 3 1999

Publication series

NameProceedings - 1999 International Conference on Information Intelligence and Systems, ICIIS 1999

Other

Other1999 International Conference on Information Intelligence and Systems, ICIIS 1999
CountryUnited States
CityBethesda
Period10/31/9911/3/99

Fingerprint

Adaptive algorithms
Learning algorithms
Support vector machines
Kalman filters
Pattern recognition
Color

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Science Applications
  • Computer Vision and Pattern Recognition

Cite this

Garg, A., Cohen, I., & Huang, T. S. (1999). Adaptive learning algorithm for SVM applied to feature tracking. In Proceedings - 1999 International Conference on Information Intelligence and Systems, ICIIS 1999 (pp. 388-395). [810293] (Proceedings - 1999 International Conference on Information Intelligence and Systems, ICIIS 1999). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICIIS.1999.810293

Adaptive learning algorithm for SVM applied to feature tracking. / Garg, Ashutosh; Cohen, Ira; Huang, Thomas S.

Proceedings - 1999 International Conference on Information Intelligence and Systems, ICIIS 1999. Institute of Electrical and Electronics Engineers Inc., 1999. p. 388-395 810293 (Proceedings - 1999 International Conference on Information Intelligence and Systems, ICIIS 1999).

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

Garg, A, Cohen, I & Huang, TS 1999, Adaptive learning algorithm for SVM applied to feature tracking. in Proceedings - 1999 International Conference on Information Intelligence and Systems, ICIIS 1999., 810293, Proceedings - 1999 International Conference on Information Intelligence and Systems, ICIIS 1999, Institute of Electrical and Electronics Engineers Inc., pp. 388-395, 1999 International Conference on Information Intelligence and Systems, ICIIS 1999, Bethesda, United States, 10/31/99. https://doi.org/10.1109/ICIIS.1999.810293
Garg A, Cohen I, Huang TS. Adaptive learning algorithm for SVM applied to feature tracking. In Proceedings - 1999 International Conference on Information Intelligence and Systems, ICIIS 1999. Institute of Electrical and Electronics Engineers Inc. 1999. p. 388-395. 810293. (Proceedings - 1999 International Conference on Information Intelligence and Systems, ICIIS 1999). https://doi.org/10.1109/ICIIS.1999.810293
Garg, Ashutosh ; Cohen, Ira ; Huang, Thomas S. / Adaptive learning algorithm for SVM applied to feature tracking. Proceedings - 1999 International Conference on Information Intelligence and Systems, ICIIS 1999. Institute of Electrical and Electronics Engineers Inc., 1999. pp. 388-395 (Proceedings - 1999 International Conference on Information Intelligence and Systems, ICIIS 1999).
@inproceedings{528c255976ac4d4aba6a85fa0f6da3ce,
title = "Adaptive learning algorithm for SVM applied to feature tracking",
abstract = "The framework of support vector machines (SVM) is becoming extremely popular in the field of statistical pattern classification. In this paper we investigate a technique which couples Kalman filter closely with the SVM. The problem of object tracking can be seen as a pattern recognition problem. However, because of the dynamics, this pattern might experience some changes over time. In order to keep track of the position of the pattern and to make out the desired pattern from the background, we must have some strong continuous time model. We propose an algorithm which combines the Markov property of the Kalman filter with the strong classification capability of SVM. The whole system has been tested on real life problems and we found that with this framework we could track a particular object even in a frame which contains identical objects. The results were compared to that of obtained by color blob tracking which showed the strength of the approach.",
author = "Ashutosh Garg and Ira Cohen and Huang, {Thomas S}",
year = "1999",
month = "1",
day = "1",
doi = "10.1109/ICIIS.1999.810293",
language = "English (US)",
series = "Proceedings - 1999 International Conference on Information Intelligence and Systems, ICIIS 1999",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "388--395",
booktitle = "Proceedings - 1999 International Conference on Information Intelligence and Systems, ICIIS 1999",
address = "United States",

}

TY - GEN

T1 - Adaptive learning algorithm for SVM applied to feature tracking

AU - Garg, Ashutosh

AU - Cohen, Ira

AU - Huang, Thomas S

PY - 1999/1/1

Y1 - 1999/1/1

N2 - The framework of support vector machines (SVM) is becoming extremely popular in the field of statistical pattern classification. In this paper we investigate a technique which couples Kalman filter closely with the SVM. The problem of object tracking can be seen as a pattern recognition problem. However, because of the dynamics, this pattern might experience some changes over time. In order to keep track of the position of the pattern and to make out the desired pattern from the background, we must have some strong continuous time model. We propose an algorithm which combines the Markov property of the Kalman filter with the strong classification capability of SVM. The whole system has been tested on real life problems and we found that with this framework we could track a particular object even in a frame which contains identical objects. The results were compared to that of obtained by color blob tracking which showed the strength of the approach.

AB - The framework of support vector machines (SVM) is becoming extremely popular in the field of statistical pattern classification. In this paper we investigate a technique which couples Kalman filter closely with the SVM. The problem of object tracking can be seen as a pattern recognition problem. However, because of the dynamics, this pattern might experience some changes over time. In order to keep track of the position of the pattern and to make out the desired pattern from the background, we must have some strong continuous time model. We propose an algorithm which combines the Markov property of the Kalman filter with the strong classification capability of SVM. The whole system has been tested on real life problems and we found that with this framework we could track a particular object even in a frame which contains identical objects. The results were compared to that of obtained by color blob tracking which showed the strength of the approach.

UR - http://www.scopus.com/inward/record.url?scp=77953223561&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77953223561&partnerID=8YFLogxK

U2 - 10.1109/ICIIS.1999.810293

DO - 10.1109/ICIIS.1999.810293

M3 - Conference contribution

AN - SCOPUS:77953223561

T3 - Proceedings - 1999 International Conference on Information Intelligence and Systems, ICIIS 1999

SP - 388

EP - 395

BT - Proceedings - 1999 International Conference on Information Intelligence and Systems, ICIIS 1999

PB - Institute of Electrical and Electronics Engineers Inc.

ER -