Incremental learning for visual tracking

Jongwoo Lim, David Ross, Ruei Sung Lin, Ming Hsuan Yang

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

100 Citations (Scopus)

Abstract

Most existing tracking algorithms construct a representation of a target object prior to the tracking task starts, and utilize invariant features to handle appearance variation of the target caused by lighting, pose, and view angle change. In this paper, we present an efficient and effective online algorithm that incrementally learns and adapts a low dimensional eigenspace representation to reflect appearance changes of the target, thereby facilitating the tracking task. Furthermore, our incremental method correctly updates the sample mean and the eigenbasis, whereas existing incremental subspace update methods ignore the fact the sample mean varies over time. The tracking problem is formulated as a state inference problem within a Markov Chain Monte Carlo framework and a particle filter is incorporated for propagating sample distributions over time. Numerous experiments demonstrate the effectiveness of the proposed tracking algorithm in indoor and outdoor environments where the target objects undergo large pose and lighting changes.

Original languageEnglish
Title of host publicationAdvances in Neural Information Processing Systems 17 - Proceedings of the 2004 Conference, NIPS 2004
PublisherNeural information processing systems foundation
ISBN (Print)0262195348, 9780262195348
StatePublished - 2005 Jan 1
Event18th Annual Conference on Neural Information Processing Systems, NIPS 2004 - Vancouver, BC, Canada
Duration: 2004 Dec 132004 Dec 16

Publication series

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

Other

Other18th Annual Conference on Neural Information Processing Systems, NIPS 2004
CountryCanada
CityVancouver, BC
Period04/12/1304/12/16

Fingerprint

Lighting
Markov processes
Experiments

Cite this

Lim, J., Ross, D., Lin, R. S., & Yang, M. H. (2005). Incremental learning for visual tracking. In Advances in Neural Information Processing Systems 17 - Proceedings of the 2004 Conference, NIPS 2004 (Advances in Neural Information Processing Systems). Neural information processing systems foundation.
Lim, Jongwoo ; Ross, David ; Lin, Ruei Sung ; Yang, Ming Hsuan. / Incremental learning for visual tracking. Advances in Neural Information Processing Systems 17 - Proceedings of the 2004 Conference, NIPS 2004. Neural information processing systems foundation, 2005. (Advances in Neural Information Processing Systems).
@inproceedings{3c5803aa1bd04f32a21f068ca5cd4827,
title = "Incremental learning for visual tracking",
abstract = "Most existing tracking algorithms construct a representation of a target object prior to the tracking task starts, and utilize invariant features to handle appearance variation of the target caused by lighting, pose, and view angle change. In this paper, we present an efficient and effective online algorithm that incrementally learns and adapts a low dimensional eigenspace representation to reflect appearance changes of the target, thereby facilitating the tracking task. Furthermore, our incremental method correctly updates the sample mean and the eigenbasis, whereas existing incremental subspace update methods ignore the fact the sample mean varies over time. The tracking problem is formulated as a state inference problem within a Markov Chain Monte Carlo framework and a particle filter is incorporated for propagating sample distributions over time. Numerous experiments demonstrate the effectiveness of the proposed tracking algorithm in indoor and outdoor environments where the target objects undergo large pose and lighting changes.",
author = "Jongwoo Lim and David Ross and Lin, {Ruei Sung} and Yang, {Ming Hsuan}",
year = "2005",
month = "1",
day = "1",
language = "English",
isbn = "0262195348",
series = "Advances in Neural Information Processing Systems",
publisher = "Neural information processing systems foundation",
booktitle = "Advances in Neural Information Processing Systems 17 - Proceedings of the 2004 Conference, NIPS 2004",

}

Lim, J, Ross, D, Lin, RS & Yang, MH 2005, Incremental learning for visual tracking. in Advances in Neural Information Processing Systems 17 - Proceedings of the 2004 Conference, NIPS 2004. Advances in Neural Information Processing Systems, Neural information processing systems foundation, 18th Annual Conference on Neural Information Processing Systems, NIPS 2004, Vancouver, BC, Canada, 04/12/13.

Incremental learning for visual tracking. / Lim, Jongwoo; Ross, David; Lin, Ruei Sung; Yang, Ming Hsuan.

Advances in Neural Information Processing Systems 17 - Proceedings of the 2004 Conference, NIPS 2004. Neural information processing systems foundation, 2005. (Advances in Neural Information Processing Systems).

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

TY - GEN

T1 - Incremental learning for visual tracking

AU - Lim, Jongwoo

AU - Ross, David

AU - Lin, Ruei Sung

AU - Yang, Ming Hsuan

PY - 2005/1/1

Y1 - 2005/1/1

N2 - Most existing tracking algorithms construct a representation of a target object prior to the tracking task starts, and utilize invariant features to handle appearance variation of the target caused by lighting, pose, and view angle change. In this paper, we present an efficient and effective online algorithm that incrementally learns and adapts a low dimensional eigenspace representation to reflect appearance changes of the target, thereby facilitating the tracking task. Furthermore, our incremental method correctly updates the sample mean and the eigenbasis, whereas existing incremental subspace update methods ignore the fact the sample mean varies over time. The tracking problem is formulated as a state inference problem within a Markov Chain Monte Carlo framework and a particle filter is incorporated for propagating sample distributions over time. Numerous experiments demonstrate the effectiveness of the proposed tracking algorithm in indoor and outdoor environments where the target objects undergo large pose and lighting changes.

AB - Most existing tracking algorithms construct a representation of a target object prior to the tracking task starts, and utilize invariant features to handle appearance variation of the target caused by lighting, pose, and view angle change. In this paper, we present an efficient and effective online algorithm that incrementally learns and adapts a low dimensional eigenspace representation to reflect appearance changes of the target, thereby facilitating the tracking task. Furthermore, our incremental method correctly updates the sample mean and the eigenbasis, whereas existing incremental subspace update methods ignore the fact the sample mean varies over time. The tracking problem is formulated as a state inference problem within a Markov Chain Monte Carlo framework and a particle filter is incorporated for propagating sample distributions over time. Numerous experiments demonstrate the effectiveness of the proposed tracking algorithm in indoor and outdoor environments where the target objects undergo large pose and lighting changes.

UR - http://www.scopus.com/inward/record.url?scp=84898933760&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84898933760

SN - 0262195348

SN - 9780262195348

T3 - Advances in Neural Information Processing Systems

BT - Advances in Neural Information Processing Systems 17 - Proceedings of the 2004 Conference, NIPS 2004

PB - Neural information processing systems foundation

ER -

Lim J, Ross D, Lin RS, Yang MH. Incremental learning for visual tracking. In Advances in Neural Information Processing Systems 17 - Proceedings of the 2004 Conference, NIPS 2004. Neural information processing systems foundation. 2005. (Advances in Neural Information Processing Systems).