GestGlove: a wearable device with gesture based touchless interaction

Abstract

Wearable devices have gained significant popularity in recent years and have enabled new modes of human computer interaction. Wearables have been explored to track user movements and gestures but typically require large arm motions. In this paper we propose a glove based wearable that is designed for active users like cyclists, motorists and skiers who typically wear a glove during their activity. The GestGlove has a gesture based user interface that enables single-handed touchless interaction. Inertial and bend sensors mounted on the device track subtle wrist and finger movements which are intuitive and convenient. We use the popular approach of Dynamic Time Warping (DTW) for the classification of this multivariate time series data. A Mahalanobis distance measure for DTW is proposed that uses the DTW warp path iteratively for metric calculation, an aspect that has not been explored in prior work.