Process

In the diagram below we can see how segmentation interacts with the different parts of the process. The images from the different cameras, corresponding to different views, are sent to the segmentation module. The segmentation module uses background segmentation to eliminate the background image. The remaining data is used with the EM algorithm to classify the pixels. Once the pixels are classified, they are sent to the Hand State Estimation module. Finally, a 3D model of the hand is displayed with the configuration of the user hand.


Interaction of the segmentation module

Figure 2. Diagram showing the interaction of the segmentation module.