Active part-decomposition, shape and motion estimation of articulated objects: a physics-based approach

Ioannis A. Kakadiaris, Dimitri Metaxas, Ruzena Bajcsy

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

46 Citations (Scopus)

Abstract

We present a novel, robust, integrated approach to segmentation shape and motion estimation of articulated objects. Initially, we assume the object consists of a single part, and we fit a deformable model to the given data using our physics-based framework. As the object attains new postures, we decide based on certain criteria if and when to replace the initial model with two new models. These criteria are based on the model's state and the given data. We then fit the models to the data using a novel algorithm for assigning forces from the data to the two models, which allows partial overlap between them and determination of joint location. This approach is applied iteratively until all the object's moving parts are identified. Furthermore, we define new global deformations and we demonstrate our technique in a series of experiments, where Kalman filtering is employed to account for noise and occlusion.

Original languageEnglish (US)
Title of host publicationProceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
PublisherPubl by IEEE
Pages980-984
Number of pages5
ISBN (Print)0818658274
StatePublished - Jan 1 1994
Externally publishedYes
EventProceedings of the 1994 IEEE Computer Society Conference on Computer Vision and Pattern Recognition - Seattle, WA, USA
Duration: Jun 21 1994Jun 23 1994

Other

OtherProceedings of the 1994 IEEE Computer Society Conference on Computer Vision and Pattern Recognition
CitySeattle, WA, USA
Period6/21/946/23/94

Fingerprint

Motion estimation
Physics
Decomposition
Experiments

All Science Journal Classification (ASJC) codes

  • Software
  • Computer Vision and Pattern Recognition

Cite this

Kakadiaris, I. A., Metaxas, D., & Bajcsy, R. (1994). Active part-decomposition, shape and motion estimation of articulated objects: a physics-based approach. In Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition (pp. 980-984). Publ by IEEE.
Kakadiaris, Ioannis A. ; Metaxas, Dimitri ; Bajcsy, Ruzena. / Active part-decomposition, shape and motion estimation of articulated objects : a physics-based approach. Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Publ by IEEE, 1994. pp. 980-984
@inproceedings{bc99eb97597d4898b8da697ddd1ac6e8,
title = "Active part-decomposition, shape and motion estimation of articulated objects: a physics-based approach",
abstract = "We present a novel, robust, integrated approach to segmentation shape and motion estimation of articulated objects. Initially, we assume the object consists of a single part, and we fit a deformable model to the given data using our physics-based framework. As the object attains new postures, we decide based on certain criteria if and when to replace the initial model with two new models. These criteria are based on the model's state and the given data. We then fit the models to the data using a novel algorithm for assigning forces from the data to the two models, which allows partial overlap between them and determination of joint location. This approach is applied iteratively until all the object's moving parts are identified. Furthermore, we define new global deformations and we demonstrate our technique in a series of experiments, where Kalman filtering is employed to account for noise and occlusion.",
author = "Kakadiaris, {Ioannis A.} and Dimitri Metaxas and Ruzena Bajcsy",
year = "1994",
month = "1",
day = "1",
language = "English (US)",
isbn = "0818658274",
pages = "980--984",
booktitle = "Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition",
publisher = "Publ by IEEE",

}

Kakadiaris, IA, Metaxas, D & Bajcsy, R 1994, Active part-decomposition, shape and motion estimation of articulated objects: a physics-based approach. in Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Publ by IEEE, pp. 980-984, Proceedings of the 1994 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA, 6/21/94.

Active part-decomposition, shape and motion estimation of articulated objects : a physics-based approach. / Kakadiaris, Ioannis A.; Metaxas, Dimitri; Bajcsy, Ruzena.

Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Publ by IEEE, 1994. p. 980-984.

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

TY - GEN

T1 - Active part-decomposition, shape and motion estimation of articulated objects

T2 - a physics-based approach

AU - Kakadiaris, Ioannis A.

AU - Metaxas, Dimitri

AU - Bajcsy, Ruzena

PY - 1994/1/1

Y1 - 1994/1/1

N2 - We present a novel, robust, integrated approach to segmentation shape and motion estimation of articulated objects. Initially, we assume the object consists of a single part, and we fit a deformable model to the given data using our physics-based framework. As the object attains new postures, we decide based on certain criteria if and when to replace the initial model with two new models. These criteria are based on the model's state and the given data. We then fit the models to the data using a novel algorithm for assigning forces from the data to the two models, which allows partial overlap between them and determination of joint location. This approach is applied iteratively until all the object's moving parts are identified. Furthermore, we define new global deformations and we demonstrate our technique in a series of experiments, where Kalman filtering is employed to account for noise and occlusion.

AB - We present a novel, robust, integrated approach to segmentation shape and motion estimation of articulated objects. Initially, we assume the object consists of a single part, and we fit a deformable model to the given data using our physics-based framework. As the object attains new postures, we decide based on certain criteria if and when to replace the initial model with two new models. These criteria are based on the model's state and the given data. We then fit the models to the data using a novel algorithm for assigning forces from the data to the two models, which allows partial overlap between them and determination of joint location. This approach is applied iteratively until all the object's moving parts are identified. Furthermore, we define new global deformations and we demonstrate our technique in a series of experiments, where Kalman filtering is employed to account for noise and occlusion.

UR - http://www.scopus.com/inward/record.url?scp=0027963749&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0027963749&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0027963749

SN - 0818658274

SP - 980

EP - 984

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

PB - Publ by IEEE

ER -

Kakadiaris IA, Metaxas D, Bajcsy R. Active part-decomposition, shape and motion estimation of articulated objects: a physics-based approach. In Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Publ by IEEE. 1994. p. 980-984