Recognizing Hand Gestures with CALI

Human computer interaction techniques using hand poses are more natural to users than those that rely on devices. In this paper we describe and evaluate two techniques for hand pose recognition, based on a general library for gesture recognition, called CALI. This library was initially designed for calligraphic gesture recognition, however its usage shows that CALI is able to broaden its application field. One of the unexplored research areas for CALI is its application to hand pose recognition. Even though there are several works on the subject, they use different approaches, like Hidden Markov Models or Model-based tracking. We developed and tested a new approach to recognize hand poses taking advantage of the features obtained from CALI. To explore this approach we implemented two techniques: the first technique recognizes bare-hands using its outer contours, the second uses color marks on each fingertips to track the hand and recognize its pose. Experimental results show that both approaches present recognitions rates around 93%.