Learning a temporally invariant representation for visual tracking

Chao Ma, Xiaokang Yang, Chongyang Zhang, Ming Hsuan Yang

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

6 Citations (Scopus)

Abstract

In this paper, we propose to learn temporally invariant features from a large number of image sequences to represent objects for visual tracking. These features are trained on a convolutional neural network with temporal invariance constraints and robust to diverse motion transformations. We employ linear correlation filters to encode the appearance templates of targets and perform the tracking task by searching for the maximum responses at each frame. The learned filters are updated online and adapt to significant appearance changes during tracking. Extensive experimental results on challenging sequences show that the proposed algorithm performs favorably against state-of-the-art methods in terms of efficiency, accuracy, and robustness.

Original languageEnglish
Title of host publication2015 IEEE International Conference on Image Processing, ICIP 2015 - Proceedings
PublisherIEEE Computer Society
Pages857-861
Number of pages5
ISBN (Electronic)9781479983391
DOIs
Publication statusPublished - 2015 Dec 9
EventIEEE International Conference on Image Processing, ICIP 2015 - Quebec City, Canada
Duration: 2015 Sep 272015 Sep 30

Publication series

NameProceedings - International Conference on Image Processing, ICIP
Volume2015-December
ISSN (Print)1522-4880

Other

OtherIEEE International Conference on Image Processing, ICIP 2015
CountryCanada
CityQuebec City
Period15/9/2715/9/30

Fingerprint

Invariance
Neural networks

All Science Journal Classification (ASJC) codes

  • Software
  • Computer Vision and Pattern Recognition
  • Signal Processing

Cite this

Ma, C., Yang, X., Zhang, C., & Yang, M. H. (2015). Learning a temporally invariant representation for visual tracking. In 2015 IEEE International Conference on Image Processing, ICIP 2015 - Proceedings (pp. 857-861). [7350921] (Proceedings - International Conference on Image Processing, ICIP; Vol. 2015-December). IEEE Computer Society. https://doi.org/10.1109/ICIP.2015.7350921
Ma, Chao ; Yang, Xiaokang ; Zhang, Chongyang ; Yang, Ming Hsuan. / Learning a temporally invariant representation for visual tracking. 2015 IEEE International Conference on Image Processing, ICIP 2015 - Proceedings. IEEE Computer Society, 2015. pp. 857-861 (Proceedings - International Conference on Image Processing, ICIP).
@inproceedings{3a412c73fa484769a7dcc4b1ea58bc22,
title = "Learning a temporally invariant representation for visual tracking",
abstract = "In this paper, we propose to learn temporally invariant features from a large number of image sequences to represent objects for visual tracking. These features are trained on a convolutional neural network with temporal invariance constraints and robust to diverse motion transformations. We employ linear correlation filters to encode the appearance templates of targets and perform the tracking task by searching for the maximum responses at each frame. The learned filters are updated online and adapt to significant appearance changes during tracking. Extensive experimental results on challenging sequences show that the proposed algorithm performs favorably against state-of-the-art methods in terms of efficiency, accuracy, and robustness.",
author = "Chao Ma and Xiaokang Yang and Chongyang Zhang and Yang, {Ming Hsuan}",
year = "2015",
month = "12",
day = "9",
doi = "10.1109/ICIP.2015.7350921",
language = "English",
series = "Proceedings - International Conference on Image Processing, ICIP",
publisher = "IEEE Computer Society",
pages = "857--861",
booktitle = "2015 IEEE International Conference on Image Processing, ICIP 2015 - Proceedings",
address = "United States",

}

Ma, C, Yang, X, Zhang, C & Yang, MH 2015, Learning a temporally invariant representation for visual tracking. in 2015 IEEE International Conference on Image Processing, ICIP 2015 - Proceedings., 7350921, Proceedings - International Conference on Image Processing, ICIP, vol. 2015-December, IEEE Computer Society, pp. 857-861, IEEE International Conference on Image Processing, ICIP 2015, Quebec City, Canada, 15/9/27. https://doi.org/10.1109/ICIP.2015.7350921

Learning a temporally invariant representation for visual tracking. / Ma, Chao; Yang, Xiaokang; Zhang, Chongyang; Yang, Ming Hsuan.

2015 IEEE International Conference on Image Processing, ICIP 2015 - Proceedings. IEEE Computer Society, 2015. p. 857-861 7350921 (Proceedings - International Conference on Image Processing, ICIP; Vol. 2015-December).

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

TY - GEN

T1 - Learning a temporally invariant representation for visual tracking

AU - Ma, Chao

AU - Yang, Xiaokang

AU - Zhang, Chongyang

AU - Yang, Ming Hsuan

PY - 2015/12/9

Y1 - 2015/12/9

N2 - In this paper, we propose to learn temporally invariant features from a large number of image sequences to represent objects for visual tracking. These features are trained on a convolutional neural network with temporal invariance constraints and robust to diverse motion transformations. We employ linear correlation filters to encode the appearance templates of targets and perform the tracking task by searching for the maximum responses at each frame. The learned filters are updated online and adapt to significant appearance changes during tracking. Extensive experimental results on challenging sequences show that the proposed algorithm performs favorably against state-of-the-art methods in terms of efficiency, accuracy, and robustness.

AB - In this paper, we propose to learn temporally invariant features from a large number of image sequences to represent objects for visual tracking. These features are trained on a convolutional neural network with temporal invariance constraints and robust to diverse motion transformations. We employ linear correlation filters to encode the appearance templates of targets and perform the tracking task by searching for the maximum responses at each frame. The learned filters are updated online and adapt to significant appearance changes during tracking. Extensive experimental results on challenging sequences show that the proposed algorithm performs favorably against state-of-the-art methods in terms of efficiency, accuracy, and robustness.

UR - http://www.scopus.com/inward/record.url?scp=84956694428&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84956694428&partnerID=8YFLogxK

U2 - 10.1109/ICIP.2015.7350921

DO - 10.1109/ICIP.2015.7350921

M3 - Conference contribution

AN - SCOPUS:84956694428

T3 - Proceedings - International Conference on Image Processing, ICIP

SP - 857

EP - 861

BT - 2015 IEEE International Conference on Image Processing, ICIP 2015 - Proceedings

PB - IEEE Computer Society

ER -

Ma C, Yang X, Zhang C, Yang MH. Learning a temporally invariant representation for visual tracking. In 2015 IEEE International Conference on Image Processing, ICIP 2015 - Proceedings. IEEE Computer Society. 2015. p. 857-861. 7350921. (Proceedings - International Conference on Image Processing, ICIP). https://doi.org/10.1109/ICIP.2015.7350921