Please use this identifier to cite or link to this item:
|Title:||3D model-based human motion capture||Authors:||LAO WEILUN||Keywords:||human motion capture, self-calibration, human body modeling||Issue Date:||7-Sep-2005||Citation:||LAO WEILUN (2005-09-07). 3D model-based human motion capture. ScholarBank@NUS Repository.||Abstract:||In this thesis, a practical framework for a 3D model-based human motion capture system is presented. Firstly, an effective linear self-calibration method for camera focal estimation based on degenerated Kruppaa??s equations is proposed. The method is based on the assumption that only the camera's focal length is unknown and the skew factor is zero. The cameraa??s focal length can be obtained in a closed form without needing any additional motion-generated information. Secondly, a novel point correspondence-based human modeling scheme from uncalibrated images is formulated. The advantage of this method is that no priori information of or measurement on the human subject or the camera setup is required. Finally, an effective motion tracking scheme is developed using a scheme based on maximising the overlapping areas between projected 2-D silhouettes of the utilised 3-D model and the foreground segmentation maps of the subject at each camera view. The approaches above are evaluated using experiments to demonstrate their accuracy and reliability.||URI:||http://scholarbank.nus.edu.sg/handle/10635/17027|
|Appears in Collections:||Master's Theses (Open)|
Show full item record
Files in This Item:
|MasterThesis_Lao Weilun(HT027215E).pdf||1.36 MB||Adobe PDF|
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.