Be The Controller: A Kinect Tool Kit for Video Game Control - Recognition of Human Motion Using Skeletal Relational Angles

Hadjiminas, N. & Child, C. H. T. (2012). Be The Controller: A Kinect Tool Kit for Video Game Control - Recognition of Human Motion Using Skeletal Relational Angles. Paper presented at the 5th Annual International Conference On Computer Games, Multimedia And Allied Technology (CGAT 2012), 2012, Bali, Indonesia.

[img]
Preview
PDF
Download (424kB) | Preview

Abstract

As technology evolves, more interactive video game controllers are being developed in order to provide players with greater interaction with games. Motion control is one of the most challenging and exciting topics in today's games industry and the related controllers have taken the industry by storm becoming a prerequisite for new generation game consoles. We propose a new method to quickly and accurately recognize human motion using skeletal relation angle recognition rather than body parts Cartesian co-ordinates position recognition in order to provide a more flexible, accurate and efficient way tracking human motion. This method was used to develop a tool kit that aims to help game designers easily identify and recognise a user’s pose and gestures using Microsoft’s Kinect motion controller, in order to then link these movements with actions in a game for example key presses can be liked to poses or gestures in order to produce key events. Finally, we attempted to evaluate the usability of the tool kit and the success rate of skeletal relation angle recognition through an experiment. Ten representative users were selected and were assigned to complete a set of tasks using the “Be the Controller” project and a game (World of Warcraft) in order to simulate real conditions of use of the “Be the Controller” project and evaluate its usability. The data from this experiment helped to form some important conclusions : users found it easy to create their own poses and gestures and they were enthusiastic with the fact that they were able to bind their action to key/mouse events and they were very satisfy with the success rate of recognitions of our method

Item Type: Conference or Workshop Item (Paper)
Uncontrolled Keywords: Pose and gesture recognition, Kinect, Be the controller tool kit, human motion, motion control
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Divisions: School of Informatics > Department of Computing
URI: http://openaccess.city.ac.uk/id/eprint/2996

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year

View more statistics