Image driven generation of pose hypotheses for 3D model-based tracking

Martim Brandão, Alexandre Bernardino, José Santos-Victor

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

4 Citations (Scopus)

Abstract

Tracking an object's 3D position and orientation from a color image can been accomplished with particle filters if its color and shape properties are known. Unfortunately, initialization in particle filters is often manual or random, thus rendering the tracking recovery process slow or no longer autonomous. A method that uses image data to generate likely pose hypotheses for known objects is proposed. These generated pose hypotheses are then used to guide visual attention and computer resources in a "top-down" tracking system such as a particle filter: speeding up the tracking process and making it more robust to unpredictable movement.

Original languageEnglish
Title of host publicationProceedings of the 12th IAPR Conference on Machine Vision Applications, MVA 2011
Pages59-62
Number of pages4
Publication statusPublished - 2011
Externally publishedYes
Event12th IAPR Conference on Machine Vision Applications, MVA 2011 - Nara
Duration: 2011 Jun 132011 Jun 15

Other

Other12th IAPR Conference on Machine Vision Applications, MVA 2011
CityNara
Period11/6/1311/6/15

Fingerprint

Color
Recovery

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition

Cite this

Brandão, M., Bernardino, A., & Santos-Victor, J. (2011). Image driven generation of pose hypotheses for 3D model-based tracking. In Proceedings of the 12th IAPR Conference on Machine Vision Applications, MVA 2011 (pp. 59-62)

Image driven generation of pose hypotheses for 3D model-based tracking. / Brandão, Martim; Bernardino, Alexandre; Santos-Victor, José.

Proceedings of the 12th IAPR Conference on Machine Vision Applications, MVA 2011. 2011. p. 59-62.

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

Brandão, M, Bernardino, A & Santos-Victor, J 2011, Image driven generation of pose hypotheses for 3D model-based tracking. in Proceedings of the 12th IAPR Conference on Machine Vision Applications, MVA 2011. pp. 59-62, 12th IAPR Conference on Machine Vision Applications, MVA 2011, Nara, 11/6/13.
Brandão M, Bernardino A, Santos-Victor J. Image driven generation of pose hypotheses for 3D model-based tracking. In Proceedings of the 12th IAPR Conference on Machine Vision Applications, MVA 2011. 2011. p. 59-62
Brandão, Martim ; Bernardino, Alexandre ; Santos-Victor, José. / Image driven generation of pose hypotheses for 3D model-based tracking. Proceedings of the 12th IAPR Conference on Machine Vision Applications, MVA 2011. 2011. pp. 59-62
@inproceedings{e1a4054f098940bfbc898376039ad9f9,
title = "Image driven generation of pose hypotheses for 3D model-based tracking",
abstract = "Tracking an object's 3D position and orientation from a color image can been accomplished with particle filters if its color and shape properties are known. Unfortunately, initialization in particle filters is often manual or random, thus rendering the tracking recovery process slow or no longer autonomous. A method that uses image data to generate likely pose hypotheses for known objects is proposed. These generated pose hypotheses are then used to guide visual attention and computer resources in a {"}top-down{"} tracking system such as a particle filter: speeding up the tracking process and making it more robust to unpredictable movement.",
author = "Martim Brand{\~a}o and Alexandre Bernardino and Jos{\'e} Santos-Victor",
year = "2011",
language = "English",
isbn = "9784901122115",
pages = "59--62",
booktitle = "Proceedings of the 12th IAPR Conference on Machine Vision Applications, MVA 2011",

}

TY - GEN

T1 - Image driven generation of pose hypotheses for 3D model-based tracking

AU - Brandão, Martim

AU - Bernardino, Alexandre

AU - Santos-Victor, José

PY - 2011

Y1 - 2011

N2 - Tracking an object's 3D position and orientation from a color image can been accomplished with particle filters if its color and shape properties are known. Unfortunately, initialization in particle filters is often manual or random, thus rendering the tracking recovery process slow or no longer autonomous. A method that uses image data to generate likely pose hypotheses for known objects is proposed. These generated pose hypotheses are then used to guide visual attention and computer resources in a "top-down" tracking system such as a particle filter: speeding up the tracking process and making it more robust to unpredictable movement.

AB - Tracking an object's 3D position and orientation from a color image can been accomplished with particle filters if its color and shape properties are known. Unfortunately, initialization in particle filters is often manual or random, thus rendering the tracking recovery process slow or no longer autonomous. A method that uses image data to generate likely pose hypotheses for known objects is proposed. These generated pose hypotheses are then used to guide visual attention and computer resources in a "top-down" tracking system such as a particle filter: speeding up the tracking process and making it more robust to unpredictable movement.

UR - http://www.scopus.com/inward/record.url?scp=84872580059&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84872580059&partnerID=8YFLogxK

M3 - Conference contribution

SN - 9784901122115

SP - 59

EP - 62

BT - Proceedings of the 12th IAPR Conference on Machine Vision Applications, MVA 2011

ER -