Introductory Techniques for 3-D Computer Vision
Introductory Techniques for 3-D Computer Vision
Multiple View Geometry in Computer Vision
Multiple View Geometry in Computer Vision
Twist Based Acquisition and Tracking of Animal and Human Kinematics
International Journal of Computer Vision
Self-calibration using a particular motion of camera
WSEAS Transactions on Computer Research
Testing inertial sensor performance as hands-free human-computer interface
WSEAS Transactions on Computers
Comparing four technologies for measuring postural micromovements during monitor engagement
Proceedings of the 30th European Conference on Cognitive Ergonomics
| Hi-index | 0.00 |
This paper presents a simple, inexpensive, and fast procedure for motion kinematics measurement and analysis [1,2]. System developed in our laboratory is based on a high speed industrial camera, active LED markers and a PC for handling cameras video stream and data analysis. Active markers used in this work were assembled using small, lightweight and easily available white LEDs. Smaller LEDs allow larger density of markers to be placed on a subject in motion, tracking position and orientation of all segments relevant for motion kinematic analysis. Computer vision algorithm for marker detection and tracking was developed in-house, followed by an algorithm for computing and analyzing kinematics data of human locomotion [3-5]. Procedures for camera calibration and sub pixel accuracy were also developed and integrated with the system. The accuracy and properties of our system were tested, and results were compared with the existing referent systems presently used in the field. Results of testing marker - camera properties suggest that the system could support work in larger volumes (distances from camera) and almost perpendicular rotations of marker against camera. This property allows building of a 3D kinematics tracking system with two or more cameras placed at different angels against the subject in setup. Proposed system has a few disadvantages; measurements and results that are representative in only one plane and use of battery powered active markers that could disturb subject during normal gait trial. The major advantage of our system is that it offers acceptable accuracy, high speed (up to 320Hz) and easy upgradeability at much lower price when compared with the other commercially available systems [6-8]. Further development of our system will include additional cameras for 3D marker tracking and integration with an inertial sensor for full kinematics and kinetic measurement of human movement.