Structure estimation of an articulated object by motion image analysis based on factorization method

被引:0
|
作者
Nagasaki, Takeshi [1 ]
Kawashima, Toshio [2 ]
Aoki, Yoshinao [1 ]
机构
[1] BUG Inc., Sapporo, 004-0015, Japan
[2] Faculty of Engineering, Hokkaido University, Sapporo, 060-0015, Japan
关键词
Computer simulation - Degrees of freedom (mechanics) - Mathematical models - Matrix algebra - Motion control - Object recognition - Spurious signal noise;
D O I
10.1002/scj.1064
中图分类号
学科分类号
摘要
A method for obtaining the structure of a target object from a motion image sequence observed from the motions of an articulated object or a multiple number of objects is proposed. By this technique, the motion and structure of each rigid body part are estimated and the restricting relationships between rigid bodies are estimated in two stages when estimating the structure of an object with a large number of degrees of freedom such as an articulated object. In the first stage, the structure and motion of each rigid body are estimated based on observed information obtained by grouping observed feature points for each rigid body, by using a factorization method having robustness against observation noise. In the second stage, the restricting relationships between rigid bodies are estimated by using the motion parameters of each rigid body. In this paper, the presence and absence of joints and the order numbers of the restricting relationships between rigid bodies are classified in terms of the ranks of the coefficient matrices by considering rotating joints and expressing the restricting relationships as linear simultaneous equations. The proposed method is applied to simulations and observations of the motions of the hand and the back of the hand and its effectiveness is verified. © 2001 Scripta Technica, Syst. Comp. Jpn.
引用
收藏
页码:69 / 79
相关论文
共 50 条
  • [41] A sequential factorization method for recovering shape and motion from image streams
    Morita, T
    Kanade, T
    IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1997, 19 (08) : 858 - 867
  • [42] SHAPE AND MOTION FROM IMAGE STREAMS UNDER ORTHOGRAPHY - A FACTORIZATION METHOD
    TOMASI, C
    KANADE, T
    INTERNATIONAL JOURNAL OF COMPUTER VISION, 1992, 9 (02) : 137 - 154
  • [43] Robust Structure and Motion Recovery Based on Augmented Factorization
    Wang, Guanghui
    IEEE ACCESS, 2017, 5 : 18999 - 19011
  • [44] Sequential factorization method for recovering shape and motion from image streams
    Fujitsu Lab Ltd, Kawasaki, Japan
    IEEE Trans Pattern Anal Mach Intell, 8 (858-867):
  • [45] A multi-body factorization method for motion analysis
    Costeira, J
    Kanade, T
    IMAGE UNDERSTANDING WORKSHOP, 1996 PROCEEDINGS, VOLS I AND II, 1996, : 1013 - 1025
  • [46] Real-time decentralized articulated motion analysis and object tracking from videos
    Qu, Wei
    Schonfeld, Dan
    IEEE TRANSACTIONS ON IMAGE PROCESSING, 2007, 16 (08) : 2129 - 2138
  • [47] Correction to: Lie-X: Depth Image Based Articulated Object Pose Estimation, Tracking, and Action Recognition on Lie Groups
    Chi Xu
    Lakshmi Narasimhan Govindarajan
    Yu Zhang
    James Stewart
    Zoë Bichler
    Suresh Jesuthasan
    Adam Claridge-Chang
    Ajay Sriram Mathuru
    Wenlong Tang
    Peixin Zhu
    Li Cheng
    International Journal of Computer Vision, 2018, 126 : 897 - 897
  • [48] Object Recoloring based on Intrinsic Image Estimation
    Beigpour, Shida
    van de Weijer, Joost
    2011 IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV), 2011, : 327 - 334
  • [49] Incorporating structure from motion uncertainty into image-based pose estimation
    Ludington, Ben T.
    Brown, Andrew P.
    Sheffler, Michael J.
    Taylor, Clark N.
    Berardi, Stephen
    GEOSPATIAL INFORMATICS, FUSION, AND MOTION VIDEO ANALYTICS V, 2015, 9473
  • [50] 3D object articulation and motion estimation in model-based stereoscopic videoconference image sequence analysis and coding
    Tzovaras, D
    Kompatsiaris, I
    Strintzis, MG
    SIGNAL PROCESSING-IMAGE COMMUNICATION, 1999, 14 (10) : 817 - 840