Gesture Recognition using Microsoft Kinect

3/27 Department Seminar

 

Marcel Simon

Master student in Computer Science from The University Of Jena, Germany

3:00 p.m., November 14, 2012
235 Weir Hall

Title: Gesture Recognition using Microsoft Kinect

Abstract:

The Microsoft Kinect was the first consumer grade camera that delivers depth values in addition to the color values. Since its launch in November 2010, the importance of research in the field of gesture control increased significantly. Gesture control is a natural and more convenient way for human computer interactions. The process of recognizing hand gestures can be significantly simplified if distance information is available for at each pixel in the image. In this talk, I will present the algorithm of integrating depth and color information for hand gesture recognition, followed by a demo of a gesture-controlled program, which was developed as a part of my Bachelor thesis. The program allows the control of a virtual 3D camera using three hand gestures only. Through the demo, we shall observe the efficiency and limitations of the algorithm.

Bio: 

Marcel Simon is a computer science student at the University of Jena, Germany. He finished the Bachelor of Science degree in 2011 and wrote his Bachelor thesis under supervision of Prof. Joachim Denzler and Marco Körner at the Chair of Computer Vision. The thesis presents an algorithm to distinguish different hand gestures using the Microsoft Kinect camera. In the same year, Marcel enrolled in the Master program of the University of Jena. In August 2012, he joined an exchange program of the University of Jena and the University of Mississippi to study abroad for one semester.