Supervised manifold learning based on biased distance for view invariant body pose estimation

Dongcheol Hur; Christian Wallraven; Seong Whan Lee

doi:10.1109/ICSMC.2012.6378158

Supervised manifold learning based on biased distance for view invariant body pose estimation

Dongcheol Hur, Christian Wallraven, Seong Whan Lee

Department of Brain and Cognitive Engineering

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

2 Citations (Scopus)

Abstract

In human body pose estimation, manifold learning is a useful method for reducing the dimension of 2D images and 3D body configuration data. Most commonly, body pose is estimated from silhouettes derived from images or image sequences. A major problem when applying manifold estimation, however, is its vulnerability to silhouette variation. In this paper, we propose a novel approach to solving viewpoint-induced silhouette variation by introducing biased label distances for learning manifolds that are able to represent variations in viewpoint, pose, and 3D body configuration. We demonstrate the effectiveness of the approach on a synthetic and a real-world dataset.

Original language	English
Title of host publication	Proceedings 2012 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2012
Pages	2717-2720
Number of pages	4
DOIs	https://doi.org/10.1109/ICSMC.2012.6378158
Publication status	Published - 2012
Event	2012 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2012 - Seoul, Korea, Republic of Duration: 2012 Oct 14 → 2012 Oct 17

Publication series

Name	Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics
ISSN (Print)	1062-922X

Other

Other	2012 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2012
Country/Territory	Korea, Republic of
City	Seoul
Period	12/10/14 → 12/10/17

Keywords

Biased distance
Manifold learning
Pose estimation

ASJC Scopus subject areas

Electrical and Electronic Engineering
Control and Systems Engineering
Human-Computer Interaction

Access to Document

10.1109/ICSMC.2012.6378158

Cite this

Hur, D., Wallraven, C., & Lee, S. W. (2012). Supervised manifold learning based on biased distance for view invariant body pose estimation. In Proceedings 2012 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2012 (pp. 2717-2720). [6378158] (Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics). https://doi.org/10.1109/ICSMC.2012.6378158

Supervised manifold learning based on biased distance for view invariant body pose estimation. / Hur, Dongcheol; Wallraven, Christian ; Lee, Seong Whan.
Proceedings 2012 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2012. 2012. p. 2717-2720 6378158 (Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics).

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

Hur, D, Wallraven, C & Lee, SW 2012, Supervised manifold learning based on biased distance for view invariant body pose estimation. in Proceedings 2012 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2012., 6378158, Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics, pp. 2717-2720, 2012 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2012, Seoul, Korea, Republic of, 12/10/14. https://doi.org/10.1109/ICSMC.2012.6378158

@inproceedings{fe2310dcd9bd4e63ab53275133df88cc,

title = "Supervised manifold learning based on biased distance for view invariant body pose estimation",

abstract = "In human body pose estimation, manifold learning is a useful method for reducing the dimension of 2D images and 3D body configuration data. Most commonly, body pose is estimated from silhouettes derived from images or image sequences. A major problem when applying manifold estimation, however, is its vulnerability to silhouette variation. In this paper, we propose a novel approach to solving viewpoint-induced silhouette variation by introducing biased label distances for learning manifolds that are able to represent variations in viewpoint, pose, and 3D body configuration. We demonstrate the effectiveness of the approach on a synthetic and a real-world dataset.",

keywords = "Biased distance, Manifold learning, Pose estimation",

author = "Dongcheol Hur and Christian Wallraven and Lee, {Seong Whan}",

year = "2012",

doi = "10.1109/ICSMC.2012.6378158",

language = "English",

isbn = "9781467317146",

series = "Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics",

pages = "2717--2720",

booktitle = "Proceedings 2012 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2012",

note = "2012 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2012 ; Conference date: 14-10-2012 Through 17-10-2012",

}

TY - GEN

T1 - Supervised manifold learning based on biased distance for view invariant body pose estimation

AU - Hur, Dongcheol

AU - Wallraven, Christian

AU - Lee, Seong Whan

PY - 2012

Y1 - 2012

N2 - In human body pose estimation, manifold learning is a useful method for reducing the dimension of 2D images and 3D body configuration data. Most commonly, body pose is estimated from silhouettes derived from images or image sequences. A major problem when applying manifold estimation, however, is its vulnerability to silhouette variation. In this paper, we propose a novel approach to solving viewpoint-induced silhouette variation by introducing biased label distances for learning manifolds that are able to represent variations in viewpoint, pose, and 3D body configuration. We demonstrate the effectiveness of the approach on a synthetic and a real-world dataset.

AB - In human body pose estimation, manifold learning is a useful method for reducing the dimension of 2D images and 3D body configuration data. Most commonly, body pose is estimated from silhouettes derived from images or image sequences. A major problem when applying manifold estimation, however, is its vulnerability to silhouette variation. In this paper, we propose a novel approach to solving viewpoint-induced silhouette variation by introducing biased label distances for learning manifolds that are able to represent variations in viewpoint, pose, and 3D body configuration. We demonstrate the effectiveness of the approach on a synthetic and a real-world dataset.

KW - Biased distance

KW - Manifold learning

KW - Pose estimation

UR - http://www.scopus.com/inward/record.url?scp=84872388802&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84872388802&partnerID=8YFLogxK

U2 - 10.1109/ICSMC.2012.6378158

DO - 10.1109/ICSMC.2012.6378158

M3 - Conference contribution

AN - SCOPUS:84872388802

SN - 9781467317146

T3 - Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics

SP - 2717

EP - 2720

BT - Proceedings 2012 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2012

T2 - 2012 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2012

Y2 - 14 October 2012 through 17 October 2012

ER -

Supervised manifold learning based on biased distance for view invariant body pose estimation

Abstract

Publication series

Other

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this