Interaction and control with the auxiliary of hand gesture

This paper presents method used hand gesture recognition in human-computer interaction and control. Nowadays in dataglove-driven motion capture field, researchers preprocess the raw sensor data of the glove with calibration methods for acquiring a high precision in the VR environment. But there are still alternative solutions. Some machine learning algorithms, for example the self-organizing map method, offer a powerful capacity in the case of hand gesture recognition with uncalibrated glove data. We present the reason we introduced the SOM into the recognition process and how it works. Our goal is to construct a stable and robust mechanism to recognize hand gesture and then put it into hand gesture control system as an interaction platform for virtual reality.

[1]  Yi Li,et al.  A novel dataglove calibration method , 2010, 2010 5th International Conference on Computer Science & Education.

[2]  Heni Ben Amor,et al.  Grasp synthesis from low‐dimensional probabilistic grasp models , 2008, Comput. Animat. Virtual Worlds.

[3]  Orit Shaer,et al.  Reality-based interaction: a framework for post-WIMP interfaces , 2008, CHI.

[4]  Michael Henry Heim,et al.  The metaphysics of virtual reality , 1993 .

[5]  Teuvo Kohonen,et al.  Data Management by Self-Organizing Maps , 2008, WCCI.

[6]  Teuvo Kohonen,et al.  Self-organized formation of topologically correct feature maps , 2004, Biological Cybernetics.

[7]  Jonathan Steuer,et al.  Defining virtual reality: dimensions determining telepresence , 1992 .

[8]  Heni Ben Amor,et al.  Grasp Recognition with Uncalibrated Data Gloves - A Comparison of Classification Methods , 2007, 2007 IEEE Virtual Reality Conference.

[9]  John W. Sammon,et al.  A Nonlinear Mapping for Data Structure Analysis , 1969, IEEE Transactions on Computers.

[10]  Teuvo Kohonen,et al.  The self-organizing map , 1990, Neurocomputing.

[11]  Kees C. J. Overbeeke,et al.  A natural human hand model , 2007, The Visual Computer.