How hand gestures are recognized using a dataglove

This paper presents research on the topic of glove based gesture recognition. Human Computer Interaction keeps moving toward interfaces which are more natural and intuitive to use, in comparison to traditional keyboard and mouse. In this paper how gestures can be recognized when using the dataglove as means of input is researched. The field of gesture recognition has many recognition techniques to offer. Based on personal experience with Hidden Markov Models (HMMs) in speech and language processing, it is eventually decided to use this technique to develop a recognizer. Gestures for basic interface tasks like clicking, dragging, zooming and rotating are defined. After having found a suitable HMM library, its algorithms are used to train and tune a HMM on a set of observation sequences. Eventually gestures are recognized with two HMMs parallelly connected. The possibilities of this gesture recognition technique are promising. However, future research in the form of an user evaluation is needed to further investigate this promising technique.