Presentation 2008-12-18
Human Area Segmentation in Vision-based Motion Capture System for Two People
Hiroaki EGASHIRA, Atsushi SHIMADA, Daisaku ARITA, Rin-ichiro TANIGUCHI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In a motion capture system (MCS) using multi-viewpoint images, it is effective to segment human area into each person when we estimate poses of two people. There are two approaches to segment the human area; one is using 2D image and the other is using 3D shape reconstructed from multi-viewpoint images. We have employed the approach to use the 3D shape which is represented in voxels. The main issue is to give each voxel a proper label which identifies the person. When a person does not touch another, it is easy to give a label by finding connected voxels. Otherwise, we have to find a boundary between the two. Therefore we define an energy function that express which label to be given for each voxel. We can get good segmentation result by minimizing the energy function based on a graph cut algorithm.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Vision-Based Motion Capture / Human Area Segmentation / Multi-Viewpoint Image / Graph Cut
Paper # PRMU2008-156
Date of Issue

Conference Information
Committee PRMU
Conference Date 2008/12/11(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Pattern Recognition and Media Understanding (PRMU)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Human Area Segmentation in Vision-based Motion Capture System for Two People
Sub Title (in English)
Keyword(1) Vision-Based Motion Capture
Keyword(2) Human Area Segmentation
Keyword(3) Multi-Viewpoint Image
Keyword(4) Graph Cut
1st Author's Name Hiroaki EGASHIRA
1st Author's Affiliation Department of Intelligent Systems, Kyushu University()
2nd Author's Name Atsushi SHIMADA
2nd Author's Affiliation Department of Intelligent Systems, Kyushu University
3rd Author's Name Daisaku ARITA
3rd Author's Affiliation Institute of Systems, Information Technologies and Nanotechnologies
4th Author's Name Rin-ichiro TANIGUCHI
4th Author's Affiliation Department of Intelligent Systems, Kyushu University
Date 2008-12-18
Paper # PRMU2008-156
Volume (vol) vol.108
Number (no) 363
Page pp.pp.-
#Pages 6
Date of Issue