On-the-fly object modeling while tracking

Zhaozheng Yin, Robert Collins

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

24 Citations (Scopus)

Abstract

To implement a persistent tracker, we build a set of view-dependent object appearance models adaptively and automatically while tracking an object under different viewing angles. This collection of acquired models is indexed with respect to the view sphere. The acquired models aid recovery from tracking failure due to occlusion and changing view angle. In this paper, view-dependent object appearance is represented by intensity patches around detected Harris corners. The intensity patches from a model are matched to the current frame by solving a bipartite linear assignment problem with outlier exclusion and missed inlier recovery. Based on these reliable matches, the change in object rotation, translation and scale is estimated between consecutive frames using Procrustes analysis. The experimental results show good performance using a collection of view-specific patch-based models for detection and tracking of vehicles in low-resolution airborne video.

Original languageEnglish (US)
Title of host publication2007 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR'07
DOIs
StatePublished - Oct 12 2007
Event2007 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR'07 - Minneapolis, MN, United States
Duration: Jun 17 2007Jun 22 2007

Publication series

NameProceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
ISSN (Print)1063-6919

Other

Other2007 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR'07
CountryUnited States
CityMinneapolis, MN
Period6/17/076/22/07

Fingerprint

Recovery

All Science Journal Classification (ASJC) codes

  • Software
  • Computer Vision and Pattern Recognition

Cite this

Yin, Z., & Collins, R. (2007). On-the-fly object modeling while tracking. In 2007 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR'07 [4270262] (Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition). https://doi.org/10.1109/CVPR.2007.383237
Yin, Zhaozheng ; Collins, Robert. / On-the-fly object modeling while tracking. 2007 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR'07. 2007. (Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition).
@inproceedings{50cf80784823420bb8a71a27c95ce340,
title = "On-the-fly object modeling while tracking",
abstract = "To implement a persistent tracker, we build a set of view-dependent object appearance models adaptively and automatically while tracking an object under different viewing angles. This collection of acquired models is indexed with respect to the view sphere. The acquired models aid recovery from tracking failure due to occlusion and changing view angle. In this paper, view-dependent object appearance is represented by intensity patches around detected Harris corners. The intensity patches from a model are matched to the current frame by solving a bipartite linear assignment problem with outlier exclusion and missed inlier recovery. Based on these reliable matches, the change in object rotation, translation and scale is estimated between consecutive frames using Procrustes analysis. The experimental results show good performance using a collection of view-specific patch-based models for detection and tracking of vehicles in low-resolution airborne video.",
author = "Zhaozheng Yin and Robert Collins",
year = "2007",
month = "10",
day = "12",
doi = "10.1109/CVPR.2007.383237",
language = "English (US)",
isbn = "1424411807",
series = "Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition",
booktitle = "2007 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR'07",

}

Yin, Z & Collins, R 2007, On-the-fly object modeling while tracking. in 2007 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR'07., 4270262, Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2007 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR'07, Minneapolis, MN, United States, 6/17/07. https://doi.org/10.1109/CVPR.2007.383237

On-the-fly object modeling while tracking. / Yin, Zhaozheng; Collins, Robert.

2007 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR'07. 2007. 4270262 (Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition).

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

TY - GEN

T1 - On-the-fly object modeling while tracking

AU - Yin, Zhaozheng

AU - Collins, Robert

PY - 2007/10/12

Y1 - 2007/10/12

N2 - To implement a persistent tracker, we build a set of view-dependent object appearance models adaptively and automatically while tracking an object under different viewing angles. This collection of acquired models is indexed with respect to the view sphere. The acquired models aid recovery from tracking failure due to occlusion and changing view angle. In this paper, view-dependent object appearance is represented by intensity patches around detected Harris corners. The intensity patches from a model are matched to the current frame by solving a bipartite linear assignment problem with outlier exclusion and missed inlier recovery. Based on these reliable matches, the change in object rotation, translation and scale is estimated between consecutive frames using Procrustes analysis. The experimental results show good performance using a collection of view-specific patch-based models for detection and tracking of vehicles in low-resolution airborne video.

AB - To implement a persistent tracker, we build a set of view-dependent object appearance models adaptively and automatically while tracking an object under different viewing angles. This collection of acquired models is indexed with respect to the view sphere. The acquired models aid recovery from tracking failure due to occlusion and changing view angle. In this paper, view-dependent object appearance is represented by intensity patches around detected Harris corners. The intensity patches from a model are matched to the current frame by solving a bipartite linear assignment problem with outlier exclusion and missed inlier recovery. Based on these reliable matches, the change in object rotation, translation and scale is estimated between consecutive frames using Procrustes analysis. The experimental results show good performance using a collection of view-specific patch-based models for detection and tracking of vehicles in low-resolution airborne video.

UR - http://www.scopus.com/inward/record.url?scp=35148853652&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=35148853652&partnerID=8YFLogxK

U2 - 10.1109/CVPR.2007.383237

DO - 10.1109/CVPR.2007.383237

M3 - Conference contribution

AN - SCOPUS:35148853652

SN - 1424411807

SN - 9781424411801

T3 - Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition

BT - 2007 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR'07

ER -

Yin Z, Collins R. On-the-fly object modeling while tracking. In 2007 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR'07. 2007. 4270262. (Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition). https://doi.org/10.1109/CVPR.2007.383237