Extracting human limb region using optical flow and nonlinear optimization

Proceedings of the Fifth Asian Conference on Computer Vision : ACCV 2002 Volume 2 Page 593-597 published_at 2002-01
アクセス数 : 846
ダウンロード数 : 111

今月のアクセス数 : 0
今月のダウンロード数 : 0
File
accv2002_limb.pdf 453 KB 種類 : fulltext
Title ( eng )
Extracting human limb region using optical flow and nonlinear optimization
Creator
Yamamura Tsuyoshi
Ohnishi Noboru
Source Title
Proceedings of the Fifth Asian Conference on Computer Vision : ACCV 2002
Volume 2
Start Page 593
End Page 597
Abstract
We propose a method for extracting human limb regions by combination of optical flow based motion segmentation and nonlinear optimization based image registration. First, rotating limb regions with rough boundaries are extracted and motion parameters are estimated for an approximated model. Then the extracted region and estimated parameters are used as initial values for nonlinear optimization that minimizes residuals of two successive frames and estimates motion parameters. Combining the two steps reduces computational costs and avoids the initial state problem of optimization. According to estimated parameters, the limb region is extracted by Bayesian classifier to obtain accurate region boundaries. Experimental results on real images will be shown.
NDC
Electrical engineering [ 540 ]
Language
eng
Resource Type conference paper
Publisher
Asian Federation of Computer Vision Societies
Date of Issued 2002-01
Rights
Copyright (c) 2002 Asian Federation of Computer Vision Societies
Publish Type Version of Record
Access Rights open access
Source Identifier
[URI] http://ir.lib.hiroshima-u.ac.jp/00020426