<-上一篇/Previous Article 下一篇/Next Article->

[1]张元元,吴晓娟,李秀媛,等.平行线约束下的视角无关步态识别算法[J].智能系统学报,2009,4(3):264-269.
　ZHANG Yuan-yuan,WU Xiao-juan,LI Xiu-yuan,et al.Viewpointindependent gait recognition with parallel line constraints[J].CAAI Transactions on Intelligent Systems,2009,4(3):264-269.

点击复制

平行线约束下的视角无关步态识别算法

PDF下载 HTML

《智能系统学报》[ISSN 1673-4785/CN 23-1538/TP] 卷: 4 期数: 2009年第3期页码: 264-269 栏目: 学术论文—机器感知与模式识别出版日期: 2009-06-25

Title:: Viewpointindependent gait recognition with parallel line constraints

文章编号:: 1673-4785(2009)03-0264-06

作者:: 张元元¹, 吴晓娟^1, 李秀媛¹, 阮秋琦²; 1.山东大学信息科学与工程学院，山东济南 250100；
2．北京交通大学信息科学研究所，北京 100044

Author(s):: ZHANG Yuan-yuan¹, WU Xiao-juan¹, LI Xiu-yuan¹, RUAN Qiu-qi^2; 1. School of Information Science and Technology, Shandong University, Ji’nan 250100, China;
2. Institute of Information Science, Beijing Jiaotong University, Beijing 100044, China

关键词:: 单目摄像机; 平行线约束; 三维重建; 视角无关; 步态识别

Keywords:: monocular camera; parallel restriction; 3D reconstruction; viewpoint independent; gait recognition

分类号:: TP391.4

文献标志码:: A

摘要:: 提出了一种满足一定约束条件的与视角无关的步态识别算法.首先给出了与视角无关步态特征的定义及约束条件，进而探讨了在单目平行线约束下空间点的坐标重建方法，利用相应的坐标转换因子可以从拍摄到的二维图像恢复出关键点的空间三维坐标.然后将人体建模成一个相互连接的三棍模型，利用这种坐标重建方法可以恢复出模型的参数，并定义了由模型参数表示的步态特征向量，即与视角无关的步态特征.理论推导和实验结果表明，这种方法在理想情况下能克服视角因素的影响.虽然得到的正确识别率不高，但它提供了多种视角交叉进行识别的可能性.

Abstract:: This paper proposes a novel gait recognition algorithm that is independent of viewpoints under certain constraints. First, we described the definition of the proposed gait feature and its constraints. Then we discussed a coordinate reconstruction method for spatial points under the constraints of monocular parallel lines. With this we could employ the relevant coordinate conversion factor to recover 3D coordinates of some key points from 2D monocular camera images. The human body was modeled and simplified as three connected sticks, and the parameters of that model were estimated using the proposed reconstruction method. Thus, we obtained viewpointindependent gait features represented by those parameters. Both theoretical calculations and experimental results revealed that the proposed gait features partially avoid the influence of viewpoint under ideal circumstances. Though correct classification rates are not high enough, it provides a useful tool for the identification of human gaits at arbitrary viewing angles.

参考文献/References:: ［1］SARKAR S, PHILLIPS P J, LIU Z, et al. The humanID gait challenge problem: data sets, performance, and analysis［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2005, 27(2): 162177.
［2］KIRBY M, SIROVICH L. Application of the KL procedure for the characterization of human faces［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1990, 12(1): 103108.
［3］HUANG P S, HARRIS C J, NIXON M S. Recognizing humans by gait via parametric canonical space［J］. Journal of Artificial Intelligence in Engineering, 1999, 13(4): 359366.
［4］王亮, 胡卫明, 谭铁牛. 基于步态的身份识别［J］．计算机学报, 2003, 26(3): 353360.
WANG Liang, HU Weiming, TAN Tieniu. Gaitbased human identification［J］. Chinese Journal of Computer, 2003, 26(3): 353360.
［5］WANG Liang, TAN Tieniu, HU Weiming, et al. Automatic gait recognition based on statistical shape analysis［J］. IEEE Transaction on Image Processing, 2003, 12(9): 11201131
［6］YAM C Y, NIXON M S, CARTER N J. Gait recognition by walking and running: a modelbased approach［C］//Proceedings of the 5th Asian Conference on Computer Vision. Melbourne, Australia, 2002: 2325.
［7］SHAKHNAROVICH G, LEE L, DARRELL T. Integrated face and gait recognition from multiple views［C］//Proceedings of CS Conference on Computer Vision and Pattern Recognition. Kauai： Hawaii, USA, 2001, 1: 439446.
?［8］URTASUN R, FUA P. 3D tracking for gait characterization and recognition［C］//Proceedings of the Sixth IEEE International Conference on Automatic Face and Gesture Recognition. Seoul, Korea, 2004: 1722
［9］AVIDAN S, SHASHUA A. Trajectory triangulation: 3D reconstruction of moving points from a monocular image sequence［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2000, 22(4): 348357.
［10］PONSA D, LOPEZ A, LUMBRERAS F, et al. 3D vehicle sensor based on monocular vision［C］// Proceedings of IEEE Intelligent Transportation Systems. Vienna, Austria, 2005: 10961101. 
［11］生物识别与安全技术研究中心. NLPR步态数据库［EB/OL］. (20011210)［20080601］. http://www.sinobiometrics.com/china/Gait%20Databases%20CH.asp. Center for Biometrics and Security Research. NLPR Database［EB/OL］. (20011210)［20080601］. http://www.sinobiometrics.com/english/Gait%20Databases.asp.
［12］QI Lei, WU Xiaojuan, ZHANG Yuanyuan, et al. New method of 3D point reconstruction from monocular camera［C］// MAYBANK S J, DING Mingyue, WAHL F, et al. MIPPR (Multispectral Image Processing and Pattern Recognition) 2007: Pattern Recognition and Computer Vision. Bellingham, USA: SPIE, 2007, 6788: 688808.

相似文献/References:: [1]张元元,吴晓娟,李秀媛,等.单目平行线约束下的空间点坐标恢复[J].智能系统学报,2010,5(1):24.
　ZHANG Yuan-yuan,WU Xiao-juan,LI Xiu-yuan,et al.Extraction of 3D coordinates using parallel lines in monocular images[J].CAAI Transactions on Intelligent Systems,2010,5():24.

备注/Memo

收稿日期：2008-06-03.
基金项目：国家自然科学基金资助项目（60675024）
通信作者：吴晓娟.E-mail:xiaojwu@sdu.edu.cn.
作者简介：
张元元，男，1984年生，博士研究生，主要研究方向计算机视觉、图像处理和模式识别等.
吴晓娟，女，1944年生，教授，博士生导师，山东大学信息科学与工程学院图像处理与模式识别方向的学科带头人.主要研究方向为智能信息处理、图像处理、模式识别、计算机视觉等.近几年来,主持了国家自然科学基金、军工、国际合作、教育部、省自然科学基金、省科技攻关等项目20余项.发表学术论文100余篇，其中有近30篇被SCI、EI检索.
李秀媛，女，1972年生，博士研究生，主要研究方向为无线通信、计算机网络等.

更新日期/Last Update: 2009-09-14

平行线约束下的视角无关步态识别算法 PDF下载HTML

备注/Memo

平行线约束下的视角无关步态识别算法

PDF下载 HTML