Recognizing articulated objects with information theoretic methods

Davi Geiger, Tyng Luh Liu

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

Abstract

This paper addresses the problem of recognizing articulated and deformable objects. In particular we are interested in human arm and leg articulations. Our approach is a Bayesian-Information integration of shape similarity and snakes, and naturally combines top-down & bottom-up algorithms. The bottom-up method extracts edges, then constructs snakes (or contours) by grouping edge elements and feeds the shape analysis. The top-down one uses shape analysis, by comparing the object model with the extracted snakes, to guide/prune the search for other snakes. The optimizations are based on Dijkstra algorithm and further pruning of this algorithm is obtained by 'integration by parts'. Our approach is general enough to handle three dimensional objects, but our focus here is on two dimensional contours.

Original languageEnglish (US)
Title of host publicationProceedings of the International Conference on Automatic Face and Gesture Recognition
PublisherIEEE
Pages45-50
Number of pages6
StatePublished - 1996
EventProceedings of the 1996 2nd International Conference on Automatic Face and Gesture Recognition - Killington, VT, USA
Duration: Oct 14 1996Oct 16 1996

Other

OtherProceedings of the 1996 2nd International Conference on Automatic Face and Gesture Recognition
CityKillington, VT, USA
Period10/14/9610/16/96

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition

Cite this

Geiger, D., & Liu, T. L. (1996). Recognizing articulated objects with information theoretic methods. In Proceedings of the International Conference on Automatic Face and Gesture Recognition (pp. 45-50). IEEE.

Recognizing articulated objects with information theoretic methods. / Geiger, Davi; Liu, Tyng Luh.

Proceedings of the International Conference on Automatic Face and Gesture Recognition. IEEE, 1996. p. 45-50.

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

Geiger, D & Liu, TL 1996, Recognizing articulated objects with information theoretic methods. in Proceedings of the International Conference on Automatic Face and Gesture Recognition. IEEE, pp. 45-50, Proceedings of the 1996 2nd International Conference on Automatic Face and Gesture Recognition, Killington, VT, USA, 10/14/96.
Geiger D, Liu TL. Recognizing articulated objects with information theoretic methods. In Proceedings of the International Conference on Automatic Face and Gesture Recognition. IEEE. 1996. p. 45-50
Geiger, Davi ; Liu, Tyng Luh. / Recognizing articulated objects with information theoretic methods. Proceedings of the International Conference on Automatic Face and Gesture Recognition. IEEE, 1996. pp. 45-50
@inproceedings{57baecad7bba40edafc2856b349333ec,
title = "Recognizing articulated objects with information theoretic methods",
abstract = "This paper addresses the problem of recognizing articulated and deformable objects. In particular we are interested in human arm and leg articulations. Our approach is a Bayesian-Information integration of shape similarity and snakes, and naturally combines top-down & bottom-up algorithms. The bottom-up method extracts edges, then constructs snakes (or contours) by grouping edge elements and feeds the shape analysis. The top-down one uses shape analysis, by comparing the object model with the extracted snakes, to guide/prune the search for other snakes. The optimizations are based on Dijkstra algorithm and further pruning of this algorithm is obtained by 'integration by parts'. Our approach is general enough to handle three dimensional objects, but our focus here is on two dimensional contours.",
author = "Davi Geiger and Liu, {Tyng Luh}",
year = "1996",
language = "English (US)",
pages = "45--50",
booktitle = "Proceedings of the International Conference on Automatic Face and Gesture Recognition",
publisher = "IEEE",

}

TY - GEN

T1 - Recognizing articulated objects with information theoretic methods

AU - Geiger, Davi

AU - Liu, Tyng Luh

PY - 1996

Y1 - 1996

N2 - This paper addresses the problem of recognizing articulated and deformable objects. In particular we are interested in human arm and leg articulations. Our approach is a Bayesian-Information integration of shape similarity and snakes, and naturally combines top-down & bottom-up algorithms. The bottom-up method extracts edges, then constructs snakes (or contours) by grouping edge elements and feeds the shape analysis. The top-down one uses shape analysis, by comparing the object model with the extracted snakes, to guide/prune the search for other snakes. The optimizations are based on Dijkstra algorithm and further pruning of this algorithm is obtained by 'integration by parts'. Our approach is general enough to handle three dimensional objects, but our focus here is on two dimensional contours.

AB - This paper addresses the problem of recognizing articulated and deformable objects. In particular we are interested in human arm and leg articulations. Our approach is a Bayesian-Information integration of shape similarity and snakes, and naturally combines top-down & bottom-up algorithms. The bottom-up method extracts edges, then constructs snakes (or contours) by grouping edge elements and feeds the shape analysis. The top-down one uses shape analysis, by comparing the object model with the extracted snakes, to guide/prune the search for other snakes. The optimizations are based on Dijkstra algorithm and further pruning of this algorithm is obtained by 'integration by parts'. Our approach is general enough to handle three dimensional objects, but our focus here is on two dimensional contours.

UR - http://www.scopus.com/inward/record.url?scp=0030379665&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0030379665&partnerID=8YFLogxK

M3 - Conference contribution

SP - 45

EP - 50

BT - Proceedings of the International Conference on Automatic Face and Gesture Recognition

PB - IEEE

ER -