ChildCrowds: A Computer Vision based Crowd Gaming framework for children

Student thesis: Master thesis (including HD thesis)

  • Tommy Paaske Hansen
4. term, Medialogy, Master (Master Programme)
The project presented here has been an attempt at creating the foundations of a computer vision based crowd gaming framework for children through which they can engage in collaborative play.

A large part of the project is spend analyzing different concepts in relation to designing an implementing such a framework.
This include analyzing theories of crowd and group behavior, childrens play and the different stages of play, definitions of games and game genres, as well as an extensive State Of The Art review on existing crowd interaction projects.
Analysis of computer vision techniques, crowd gaming activities and how to test with a child target group is also included.

Following the analysis chapters a design of a Framework involving a modified playstation 3 camera and colour segmentation of chemical glow sticks in the video feed provided by this camera, is presented.
3 different interaction paradigms were designed around the mean positions of green and blue glow sticks, and 3 games were designed to evaluate the framework through these interaction paradigms.

Finally the framework was tested and evaluated through observation of 3 gaming sessions, ranking scale questionnaires and a focus group interview with 7 children from a after school centre.
SpecialisationGames
LanguageEnglish
Publication date30 Aug 2013
Number of pages108
ID: 80595580