With the advent of Kinect (and related technologies), minority report style scenarios where users interact with computing devices using freeform hand gestures (or other body movements) have become a reality. However, while hand gestures provide a natural way to interact with the environment, they can be difficult for a computer to understand as the movements seldom are exactly (or at least easily) repeatable [1] . The aim of this thesis is to develop and evaluate a robust gesture tracking system on top of Kinect, and to quantify the complexity and repeatability of the gestures that can be supported using the system [2].

References:

[1] Oulasvirta, A.; Roos, T.; Modig, A. & Leppänen, L. Information capacity of full-body movements Proceedings of the 2013 ACM annual conference on Human factors in computing systems (CHI), ACM, 2013, 1289-1298

[2] Ren, Z.; Meng, J.; Yuan, J. & Zhang, Z. Robust hand gesture recognition with Kinect sensor Proceedings of the 19th ACM international conference on Multimedia (MM), ACM, 2011, 759-760