For the performance evaluation of the hand tracking and gesture recognition, the system has been tested
on a set of 24 users. Each user has performed a predefined set of 40 movements and therefore we have 960
gestures to evaluate the application results. It is natural to think that the system’s accuracy will be measured
controlling the performance of the desired user movements for managing the videogame. This sequence
included all the application possible states and transitions. Figure 6 shows the performance evaluation
results. These results are represented using a bidimensional matrix with the application states as columns and
the number of appearances of the gesture as rows. The columns are paired for each gesture: the first column
is the number of tests of the gesture that has been correctly identified; the second column is the total number
of times that the gesture has been carried out. As it can be seen in Fig. 6, the hand recognition gesture works
fine for a 99% of the cases.