Real-time, 3D estimation of human body postures from trinocular images

Shoichiro Iwasawa; Jun Ohya; Kazuhiko Takahashi; Tatsumi Sakaguchi; Sinjiro Kawato; Kazuyuki Ebihara; Shigeo Morishima

doi:10.1109/PEOPLE.1999.798340

Real-time, 3D estimation of human body postures from trinocular images

Shoichiro Iwasawa, Jun Ohya, Kazuhiko Takahashi, Tatsumi Sakaguchi, Sinjiro Kawato, Kazuyuki Ebihara, Shigeo Morishima

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

31 Citations (Scopus)

Abstract

This paper proposes a new real-time method for estimating human postures in 3D from trinocular images. In this method, an upper body orientation detection and a heuristic contour analysis are performed on the human silhouettes extracted from the trinocular images so that representative points such as the top of the head can be located. The major joint positions are estimated based on a genetic algorithm based learning procedure. 3D coordinates of the representative points and joints are then obtained from the two views by evaluating the appropriateness of the three views. The proposed method implemented on a personal computer runs in real-time (30 frames/second). Experimental results show high estimation accuracies and the effectiveness of the view selection process.

Original language	English
Title of host publication	Proceedings - IEEE International Workshop on Modelling People, MPeople 1999
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	3-10
Number of pages	8
ISBN (Electronic)	0769503624, 9780769503622
DOIs	https://doi.org/10.1109/PEOPLE.1999.798340
Publication status	Published - 1999 Jan 1
Externally published	Yes
Event	1999 IEEE International Workshop on Modelling People, MPeople 1999 - Kerkyra, Greece Duration: 1999 Sept 20 → …

Other

Other	1999 IEEE International Workshop on Modelling People, MPeople 1999
Country/Territory	Greece
City	Kerkyra
Period	99/9/20 → …

ASJC Scopus subject areas

Computer Vision and Pattern Recognition
Modelling and Simulation
Human Factors and Ergonomics

Access to Document

10.1109/PEOPLE.1999.798340

Cite this

Iwasawa, S., Ohya, J., Takahashi, K., Sakaguchi, T., Kawato, S., Ebihara, K., & Morishima, S. (1999). Real-time, 3D estimation of human body postures from trinocular images. In Proceedings - IEEE International Workshop on Modelling People, MPeople 1999 (pp. 3-10). Article 798340 Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/PEOPLE.1999.798340

Real-time, 3D estimation of human body postures from trinocular images. / Iwasawa, Shoichiro; Ohya, Jun; Takahashi, Kazuhiko et al.
Proceedings - IEEE International Workshop on Modelling People, MPeople 1999. Institute of Electrical and Electronics Engineers Inc., 1999. p. 3-10 798340.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Iwasawa, S, Ohya, J, Takahashi, K, Sakaguchi, T, Kawato, S, Ebihara, K & Morishima, S 1999, Real-time, 3D estimation of human body postures from trinocular images. in Proceedings - IEEE International Workshop on Modelling People, MPeople 1999., 798340, Institute of Electrical and Electronics Engineers Inc., pp. 3-10, 1999 IEEE International Workshop on Modelling People, MPeople 1999, Kerkyra, Greece, 99/9/20. https://doi.org/10.1109/PEOPLE.1999.798340

@inproceedings{f4c59ca15e5148f38647a8cd75ce1c0e,

title = "Real-time, 3D estimation of human body postures from trinocular images",

abstract = "This paper proposes a new real-time method for estimating human postures in 3D from trinocular images. In this method, an upper body orientation detection and a heuristic contour analysis are performed on the human silhouettes extracted from the trinocular images so that representative points such as the top of the head can be located. The major joint positions are estimated based on a genetic algorithm based learning procedure. 3D coordinates of the representative points and joints are then obtained from the two views by evaluating the appropriateness of the three views. The proposed method implemented on a personal computer runs in real-time (30 frames/second). Experimental results show high estimation accuracies and the effectiveness of the view selection process.",

author = "Shoichiro Iwasawa and Jun Ohya and Kazuhiko Takahashi and Tatsumi Sakaguchi and Sinjiro Kawato and Kazuyuki Ebihara and Shigeo Morishima",

year = "1999",

month = jan,

day = "1",

doi = "10.1109/PEOPLE.1999.798340",

language = "English",

pages = "3--10",

booktitle = "Proceedings - IEEE International Workshop on Modelling People, MPeople 1999",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

note = "1999 IEEE International Workshop on Modelling People, MPeople 1999 ; Conference date: 20-09-1999",

}

TY - GEN

T1 - Real-time, 3D estimation of human body postures from trinocular images

AU - Iwasawa, Shoichiro

AU - Ohya, Jun

AU - Takahashi, Kazuhiko

AU - Sakaguchi, Tatsumi

AU - Kawato, Sinjiro

AU - Ebihara, Kazuyuki

AU - Morishima, Shigeo

PY - 1999/1/1

Y1 - 1999/1/1

N2 - This paper proposes a new real-time method for estimating human postures in 3D from trinocular images. In this method, an upper body orientation detection and a heuristic contour analysis are performed on the human silhouettes extracted from the trinocular images so that representative points such as the top of the head can be located. The major joint positions are estimated based on a genetic algorithm based learning procedure. 3D coordinates of the representative points and joints are then obtained from the two views by evaluating the appropriateness of the three views. The proposed method implemented on a personal computer runs in real-time (30 frames/second). Experimental results show high estimation accuracies and the effectiveness of the view selection process.

AB - This paper proposes a new real-time method for estimating human postures in 3D from trinocular images. In this method, an upper body orientation detection and a heuristic contour analysis are performed on the human silhouettes extracted from the trinocular images so that representative points such as the top of the head can be located. The major joint positions are estimated based on a genetic algorithm based learning procedure. 3D coordinates of the representative points and joints are then obtained from the two views by evaluating the appropriateness of the three views. The proposed method implemented on a personal computer runs in real-time (30 frames/second). Experimental results show high estimation accuracies and the effectiveness of the view selection process.

UR - http://www.scopus.com/inward/record.url?scp=85031519341&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85031519341&partnerID=8YFLogxK

U2 - 10.1109/PEOPLE.1999.798340

DO - 10.1109/PEOPLE.1999.798340

M3 - Conference contribution

AN - SCOPUS:85031519341

SP - 3

EP - 10

BT - Proceedings - IEEE International Workshop on Modelling People, MPeople 1999

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 1999 IEEE International Workshop on Modelling People, MPeople 1999

Y2 - 20 September 1999

ER -

Real-time, 3D estimation of human body postures from trinocular images

Abstract

Other

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this