Abstract
The automatic interpretation of gestures based on computer vision offers new possibilities to interact with machines. These interactions are more natural and more intuitive than those with classical devices. In this paper, we are interested in using our hands as pointing devices to remotely ordering computers. The proposed application can be summarized in three main steps: hands detection in a video, tracking of detected hands and converting hands shapes or trajectories into computer orders.
To achieve this application a classification phase is paramount whether at the part of hands detection, or at the phase of "ordering computer". For this reason, we have employed a wavelet network classifier (WNC) based on fast wavelet transform (FWT) for its robustness and for its pertinent results in the classification's domain. Experiments show that our employed methods are effective for real-time hands detection, hands tracking and hands gestures recognition.