Real-Time 3D Hand Interaction: Single Webcam Low-Cost Approach

The purpose of this project is to create a library that will allow its users to control 3D applications by using one or both of their hands. The final product could easily be incorporated into 3D applications, each customized to utilize a set of poses. Even though off-the-shelf motion capture gloves have reached lower prices in recent years, they are still expensive for home users. The algorithm suggested is based only on a single webcam combined with coded palm and fingers. Users should be able to code one or more of the fingers. One webcam is still somewhat constraining as two should ideally be used for 3D mapping of the hand, but by additionally using palm and finger coding we can greatly improve precision and, most importantly, reduce the processing power required for feasible real-time 3D interaction.

[1]  Nicol N. Schraudolph,et al.  3D hand tracking by rapid stochastic gradient descent using a skinning model , 2004 .

[2]  David J. Fleet,et al.  Stochastic Tracking of 3D Human Figures Using 2D Image Motion , 2000, ECCV.

[3]  Luc Van Gool,et al.  Fast stochastic optimization for articulated structure tracking , 2007, Image Vis. Comput..

[4]  Paulo R. S. Mendonça,et al.  Model-Based Hand Tracking Using an Unscented Kalman Filter , 2001, BMVC.

[5]  Paulo R. S. Mendonça,et al.  Model-based 3D tracking of an articulated hand , 2001, Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001.

[6]  Michael Isard,et al.  CONDENSATION—Conditional Density Propagation for Visual Tracking , 1998, International Journal of Computer Vision.

[7]  Luc Van Gool,et al.  Smart particle filtering for high-dimensional tracking , 2007, Comput. Vis. Image Underst..