Monocular video based markerless 3D human pose estimation by using local multi-connected belief propagation with multi-cue fusion

Chenguang Liu; Jiafeng Liu; Jianhua Huang; Xianglong Tang

doi:10.1109/ICICISYS.2010.5658593

Monocular video based markerless 3D human pose estimation by using local multi-connected belief propagation with multi-cue fusion

Chenguang Liu^*
, Jiafeng Liu
, Jianhua Huang
, Xianglong Tang

^*Corresponding author for this work

Faculty of Computing

Harbin Institute of Technology

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

1 Scopus citations

Abstract

In this paper, we put forward a concept of Local Multi-connected Belief Propagation (LMBP) and apply it on a 2D body model to capture 2D image coordinates of the human body joints. When performing the likelihood measurement, three different image features are fused adaptively to construct the likelihood function. And then, we lift the 2D tracking results to 3D human poses by utilizing Tylor method. The proposed method is then applied to several video sequences which contain different subjects, challenging human motions and complex backgrounds, and the tracking results indicate its effectiveness.

Original language	English
Title of host publication	Proceedings - 2010 IEEE International Conference on Intelligent Computing and Intelligent Systems, ICIS 2010
Pages	478-482
Number of pages	5
DOIs	https://doi.org/10.1109/ICICISYS.2010.5658593
State	Published - 2010
Event	2010 IEEE International Conference on Intelligent Computing and Intelligent Systems, ICIS 2010 - Xiamen, China Duration: 29 Oct 2010 → 31 Oct 2010

Publication series

Name	Proceedings - 2010 IEEE International Conference on Intelligent Computing and Intelligent Systems, ICIS 2010
Volume	1

Conference

Conference	2010 IEEE International Conference on Intelligent Computing and Intelligent Systems, ICIS 2010
Country/Territory	China
City	Xiamen
Period	29/10/10 → 31/10/10

Keywords

3D human motion tracking
Adaptive multiple features fusion
Constrained 2D human body model
Local multi-connected belie/propagation (LMBP)

Access to Document

10.1109/ICICISYS.2010.5658593

Cite this

Liu, C., Liu, J., Huang, J., & Tang, X. (2010). Monocular video based markerless 3D human pose estimation by using local multi-connected belief propagation with multi-cue fusion. In Proceedings - 2010 IEEE International Conference on Intelligent Computing and Intelligent Systems, ICIS 2010 (pp. 478-482). Article 5658593 (Proceedings - 2010 IEEE International Conference on Intelligent Computing and Intelligent Systems, ICIS 2010; Vol. 1). https://doi.org/10.1109/ICICISYS.2010.5658593

Liu, Chenguang ; Liu, Jiafeng ; Huang, Jianhua et al. / Monocular video based markerless 3D human pose estimation by using local multi-connected belief propagation with multi-cue fusion. Proceedings - 2010 IEEE International Conference on Intelligent Computing and Intelligent Systems, ICIS 2010. 2010. pp. 478-482 (Proceedings - 2010 IEEE International Conference on Intelligent Computing and Intelligent Systems, ICIS 2010).

@inproceedings{13eec38737cd417881abeff850309033,

title = "Monocular video based markerless 3D human pose estimation by using local multi-connected belief propagation with multi-cue fusion",

abstract = "In this paper, we put forward a concept of Local Multi-connected Belief Propagation (LMBP) and apply it on a 2D body model to capture 2D image coordinates of the human body joints. When performing the likelihood measurement, three different image features are fused adaptively to construct the likelihood function. And then, we lift the 2D tracking results to 3D human poses by utilizing Tylor method. The proposed method is then applied to several video sequences which contain different subjects, challenging human motions and complex backgrounds, and the tracking results indicate its effectiveness.",

keywords = "3D human motion tracking, Adaptive multiple features fusion, Constrained 2D human body model, Local multi-connected belie/propagation (LMBP)",

author = "Chenguang Liu and Jiafeng Liu and Jianhua Huang and Xianglong Tang",

year = "2010",

doi = "10.1109/ICICISYS.2010.5658593",

language = "英语",

isbn = "9781424465835",

series = "Proceedings - 2010 IEEE International Conference on Intelligent Computing and Intelligent Systems, ICIS 2010",

pages = "478--482",

booktitle = "Proceedings - 2010 IEEE International Conference on Intelligent Computing and Intelligent Systems, ICIS 2010",

note = "2010 IEEE International Conference on Intelligent Computing and Intelligent Systems, ICIS 2010 ; Conference date: 29-10-2010 Through 31-10-2010",

}

Liu, C, Liu, J, Huang, J & Tang, X 2010, Monocular video based markerless 3D human pose estimation by using local multi-connected belief propagation with multi-cue fusion. in Proceedings - 2010 IEEE International Conference on Intelligent Computing and Intelligent Systems, ICIS 2010., 5658593, Proceedings - 2010 IEEE International Conference on Intelligent Computing and Intelligent Systems, ICIS 2010, vol. 1, pp. 478-482, 2010 IEEE International Conference on Intelligent Computing and Intelligent Systems, ICIS 2010, Xiamen, China, 29/10/10. https://doi.org/10.1109/ICICISYS.2010.5658593

Monocular video based markerless 3D human pose estimation by using local multi-connected belief propagation with multi-cue fusion. / Liu, Chenguang; Liu, Jiafeng; Huang, Jianhua et al.
Proceedings - 2010 IEEE International Conference on Intelligent Computing and Intelligent Systems, ICIS 2010. 2010. p. 478-482 5658593 (Proceedings - 2010 IEEE International Conference on Intelligent Computing and Intelligent Systems, ICIS 2010; Vol. 1).

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

TY - GEN

T1 - Monocular video based markerless 3D human pose estimation by using local multi-connected belief propagation with multi-cue fusion

AU - Liu, Chenguang

AU - Liu, Jiafeng

AU - Huang, Jianhua

AU - Tang, Xianglong

PY - 2010

Y1 - 2010

N2 - In this paper, we put forward a concept of Local Multi-connected Belief Propagation (LMBP) and apply it on a 2D body model to capture 2D image coordinates of the human body joints. When performing the likelihood measurement, three different image features are fused adaptively to construct the likelihood function. And then, we lift the 2D tracking results to 3D human poses by utilizing Tylor method. The proposed method is then applied to several video sequences which contain different subjects, challenging human motions and complex backgrounds, and the tracking results indicate its effectiveness.

AB - In this paper, we put forward a concept of Local Multi-connected Belief Propagation (LMBP) and apply it on a 2D body model to capture 2D image coordinates of the human body joints. When performing the likelihood measurement, three different image features are fused adaptively to construct the likelihood function. And then, we lift the 2D tracking results to 3D human poses by utilizing Tylor method. The proposed method is then applied to several video sequences which contain different subjects, challenging human motions and complex backgrounds, and the tracking results indicate its effectiveness.

KW - 3D human motion tracking

KW - Adaptive multiple features fusion

KW - Constrained 2D human body model

KW - Local multi-connected belie/propagation (LMBP)

UR - https://www.scopus.com/pages/publications/78651290759

U2 - 10.1109/ICICISYS.2010.5658593

DO - 10.1109/ICICISYS.2010.5658593

M3 - 会议稿件

AN - SCOPUS:78651290759

SN - 9781424465835

T3 - Proceedings - 2010 IEEE International Conference on Intelligent Computing and Intelligent Systems, ICIS 2010

SP - 478

EP - 482

BT - Proceedings - 2010 IEEE International Conference on Intelligent Computing and Intelligent Systems, ICIS 2010

T2 - 2010 IEEE International Conference on Intelligent Computing and Intelligent Systems, ICIS 2010

Y2 - 29 October 2010 through 31 October 2010

ER -

Liu C, Liu J, Huang J, Tang X. Monocular video based markerless 3D human pose estimation by using local multi-connected belief propagation with multi-cue fusion. In Proceedings - 2010 IEEE International Conference on Intelligent Computing and Intelligent Systems, ICIS 2010. 2010. p. 478-482. 5658593. (Proceedings - 2010 IEEE International Conference on Intelligent Computing and Intelligent Systems, ICIS 2010). doi: 10.1109/ICICISYS.2010.5658593

Monocular video based markerless 3D human pose estimation by using local multi-connected belief propagation with multi-cue fusion

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this