Improving in-game gesture learning with visual feedback

This paper presents a research work on gesture recognition and feedback to reduce the learning time of new gestures and to augment user performance in a game application. A Wiimote controlled space shooter game, GeStar Wars, has been developed. The player controls a spaceship through the buttons in the controller, while forearm gestures can be used to perform special actions. Gesture strokes are mapped in a 3x3 grid and are differentiated according to the path of the covered grid cells. In-game visual feedback displays to the user the current gesture path and which cells were covered after the gesture is performed. The novelty of this research resides in the correlated gesture recognition methodology and feedback which helps the user to learn and correct the gestures. The evaluation, conducted with 12 users, showed that the users performed significantly better if feedback was provided