The current paper is a self-laboratory documentation from Department of Control Engineering and Information Technology at Budapest University of Technology and Economics in the topic of „Human skeleton model fitting to camera stream using mobile devices”. In this paper I present the implementation of a human skeleton fitting system on a mobile platform as well as its potential applications. In the design phase I break down the problem into smaller units then I explore the solutions individually, considering modern image processing techniques.
The system was developed on the iOS operating system using Objective-C and C++ programming languages (including OpenCV, OpenTLD, cvBlobsLib) primarly focusing on real-time visualization. The application can be used without any calib- ration. After it finds a face in the video stream it starts the skeleton fitting and displays the skeleton on top of the original image. At the end of the paper I present the strength and weaknesses of the implemented system and pinpoint the possible areas where the application can be developed futher.