首頁 > 網路資源 > 大同大學數位論文系統

Title page for etd-0812114-191515


URN etd-0812114-191515 Statistics This thesis had been viewed 2122 times. Download 1852 times.
Author Chih-Yuan Hsu
Author's Email Address No Public.
Department Computer Science and Enginerring
Year 2013 Semester 2
Degree Master Type of Document Master's Thesis
Language zh-TW.Big5 Chinese Page Count 72
Title Human Motion Capture and Mistake Correction Using Kinect Camera
Keyword
  • inverse kinematics
  • motion capture
  • Kinect
  • OpenNI
  • self-occlusion
  • self-occlusion
  • OpenNI
  • Kinect
  • motion capture
  • inverse kinematics
  • Abstract Human motion capture has been an important research topic in computer vision and computer graphics. It is used in military, entertainment, sports, and medical applications. The traditional motion capture equipment are not affordable for ordinary people. With the advent of Kinect, low-cost human motion analysis and capture becomes possible.
    In this paper, using Kinect and OpenNI making human motion capture system. This solves the problem of motion capture equipment costs. Use four algorithms to carry on judging and correction error motion (such as self-occlusion and image noise...etc. produced of the error capture). But these four algorithms can not act on real-time environment. Therefore the development predicts algorithms way to carry out real-time correction effect.
    In four methods, damped least squares inverse kinematics algorithm differs from previous correction methods. Compared with the other methods, damped least squares algorithm corrects the result could be better.
    Advisor Committee
  • Jyun-Ming Chen - advisor
  • Chun-Fong You - co-chair
  • Shang-Lin Hsieh - co-chair
  • Files indicate access worldwide
    Date of Defense 2014-07-28 Date of Submission 2014-08-12


    Browse | Search All Available ETDs