Computational animation of a human skeletal model

Modern Motion Capture (MoCap) technologies utilise sophisticated equipment and advanced software to transfer human facial expressions and body movements onto digital models on a computer screen, producing state-of-the-art 3D animations of the human body. While these systems may represent the forefro...

Full description

Saved in:
Bibliographic Details
Main Author: Chin, Gabriel Peter Kehan
Other Authors: Lee Yong Tsui
Format: Final Year Project
Language:English
Published: 2017
Subjects:
Online Access:http://hdl.handle.net/10356/71866
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
Description
Summary:Modern Motion Capture (MoCap) technologies utilise sophisticated equipment and advanced software to transfer human facial expressions and body movements onto digital models on a computer screen, producing state-of-the-art 3D animations of the human body. While these systems may represent the forefront of innovation in a great number of different industries, they often require the use of multiple cameras and/or expensive bodysuits to record movements. Thus, the utilisation of MoCap technologies remains naturally more exclusive to high-budget professionals, animators and movie directors. This project therefore looks at developing the animation capabilities of a new MoCap system which is based on a single video camera, with the potential of reaching out to the everyday user. In such a system, one important aspect of development involves the design and initialisation of an appropriate 3D model that provides accurate representation of the skeletal system. Additionally, GLUI and GLUT functionalities used in conjunction with animation algorithms allow the 3D model to simulate human movement. The results of this project exhibit a basic bones system capable of displaying a walking animation on a GLUI window in two different perspectives. Survey results gathered indicate that the GLUI interface and animations generated by the source code are intuitive and adequately realistic, respectively. While the bone structure and animation are only able to demonstrate limited levels of realism, the project lays the groundwork for the future development work of the standard MoCap system that animates based on pose estimation and interpolation.