Visual tracking with online multiple instance learning

Boris Babenko, Serge Belongie, Ming Hsuan Yang

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

1416 Citations (Scopus)

Abstract

In this paper, we address the problem of learning an adaptive appearance model for object tracking. In particular, a class of tracking techniques called "tracking by detection" have been shown to give promising results at realtime speeds. These methods train a discriminative classifier in an online manner to separate the object from the background. This classifier bootstraps itself by using the current tracker state to extract positive and negative examples from the current frame. Slight inaccuracies in the tracker can therefore lead to incorrectly labeled training examples, which degrades the classifier and can cause further drift. In this paper we show that using Multiple Instance Learning (MIL) instead of traditional supervised learning avoids these problems, and can therefore lead to a more robust tracker with fewer parameter tweaks. We present a novel online MIL algorithm for object tracking that achieves superior results with real-time performance.

Original languageEnglish
Title of host publication2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPR Workshops 2009
PublisherIEEE Computer Society
Pages983-990
Number of pages8
ISBN (Print)9781424439935
DOIs
Publication statusPublished - 2009 Jan 1
Event2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition - Miami, FL, United States
Duration: 2009 Jun 202009 Jun 25

Publication series

Name2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPR Workshops 2009
Volume2009 IEEE Computer Society Conference on Computer Vision and ...

Conference

Conference2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition
CountryUnited States
CityMiami, FL
Period09/6/2009/6/25

Fingerprint

Classifiers
Supervised learning
Learning algorithms

All Science Journal Classification (ASJC) codes

  • Computer Vision and Pattern Recognition
  • Biomedical Engineering

Cite this

Babenko, B., Belongie, S., & Yang, M. H. (2009). Visual tracking with online multiple instance learning. In 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPR Workshops 2009 (pp. 983-990). [5206737] (2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPR Workshops 2009; Vol. 2009 IEEE Computer Society Conference on Computer Vision and ...). IEEE Computer Society. https://doi.org/10.1109/CVPRW.2009.5206737
Babenko, Boris ; Belongie, Serge ; Yang, Ming Hsuan. / Visual tracking with online multiple instance learning. 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPR Workshops 2009. IEEE Computer Society, 2009. pp. 983-990 (2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPR Workshops 2009).
@inproceedings{75815e09f3114894b095eecfcfc23fcb,
title = "Visual tracking with online multiple instance learning",
abstract = "In this paper, we address the problem of learning an adaptive appearance model for object tracking. In particular, a class of tracking techniques called {"}tracking by detection{"} have been shown to give promising results at realtime speeds. These methods train a discriminative classifier in an online manner to separate the object from the background. This classifier bootstraps itself by using the current tracker state to extract positive and negative examples from the current frame. Slight inaccuracies in the tracker can therefore lead to incorrectly labeled training examples, which degrades the classifier and can cause further drift. In this paper we show that using Multiple Instance Learning (MIL) instead of traditional supervised learning avoids these problems, and can therefore lead to a more robust tracker with fewer parameter tweaks. We present a novel online MIL algorithm for object tracking that achieves superior results with real-time performance.",
author = "Boris Babenko and Serge Belongie and Yang, {Ming Hsuan}",
year = "2009",
month = "1",
day = "1",
doi = "10.1109/CVPRW.2009.5206737",
language = "English",
isbn = "9781424439935",
series = "2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPR Workshops 2009",
publisher = "IEEE Computer Society",
pages = "983--990",
booktitle = "2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPR Workshops 2009",
address = "United States",

}

Babenko, B, Belongie, S & Yang, MH 2009, Visual tracking with online multiple instance learning. in 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPR Workshops 2009., 5206737, 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPR Workshops 2009, vol. 2009 IEEE Computer Society Conference on Computer Vision and ..., IEEE Computer Society, pp. 983-990, 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Miami, FL, United States, 09/6/20. https://doi.org/10.1109/CVPRW.2009.5206737

Visual tracking with online multiple instance learning. / Babenko, Boris; Belongie, Serge; Yang, Ming Hsuan.

2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPR Workshops 2009. IEEE Computer Society, 2009. p. 983-990 5206737 (2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPR Workshops 2009; Vol. 2009 IEEE Computer Society Conference on Computer Vision and ...).

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

TY - GEN

T1 - Visual tracking with online multiple instance learning

AU - Babenko, Boris

AU - Belongie, Serge

AU - Yang, Ming Hsuan

PY - 2009/1/1

Y1 - 2009/1/1

N2 - In this paper, we address the problem of learning an adaptive appearance model for object tracking. In particular, a class of tracking techniques called "tracking by detection" have been shown to give promising results at realtime speeds. These methods train a discriminative classifier in an online manner to separate the object from the background. This classifier bootstraps itself by using the current tracker state to extract positive and negative examples from the current frame. Slight inaccuracies in the tracker can therefore lead to incorrectly labeled training examples, which degrades the classifier and can cause further drift. In this paper we show that using Multiple Instance Learning (MIL) instead of traditional supervised learning avoids these problems, and can therefore lead to a more robust tracker with fewer parameter tweaks. We present a novel online MIL algorithm for object tracking that achieves superior results with real-time performance.

AB - In this paper, we address the problem of learning an adaptive appearance model for object tracking. In particular, a class of tracking techniques called "tracking by detection" have been shown to give promising results at realtime speeds. These methods train a discriminative classifier in an online manner to separate the object from the background. This classifier bootstraps itself by using the current tracker state to extract positive and negative examples from the current frame. Slight inaccuracies in the tracker can therefore lead to incorrectly labeled training examples, which degrades the classifier and can cause further drift. In this paper we show that using Multiple Instance Learning (MIL) instead of traditional supervised learning avoids these problems, and can therefore lead to a more robust tracker with fewer parameter tweaks. We present a novel online MIL algorithm for object tracking that achieves superior results with real-time performance.

UR - http://www.scopus.com/inward/record.url?scp=70450188146&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70450188146&partnerID=8YFLogxK

U2 - 10.1109/CVPRW.2009.5206737

DO - 10.1109/CVPRW.2009.5206737

M3 - Conference contribution

AN - SCOPUS:70450188146

SN - 9781424439935

T3 - 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPR Workshops 2009

SP - 983

EP - 990

BT - 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPR Workshops 2009

PB - IEEE Computer Society

ER -

Babenko B, Belongie S, Yang MH. Visual tracking with online multiple instance learning. In 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPR Workshops 2009. IEEE Computer Society. 2009. p. 983-990. 5206737. (2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPR Workshops 2009). https://doi.org/10.1109/CVPRW.2009.5206737