Basic Daily Activity Recognition with a Data Glove

Abstract

Many people in the world are affected by the Alzheimer disease leading to the dysfunctionality of the hand. In one side, this symptom is not the most important of this disease and not much attention is given to this one. In the other side, the literrature provides two main solutions such as computer vision and data glove allowing to recognize hand gestures for virtual reality or robotic applications. From this finding and need, we decided to developed our own data glove prototype allowing to monitor the evolution of the dysfunctionality of the hand by recognizing objects in basic daily activities. Our approach is simple, cheap (~220$) and efficient (~100% of correct predictions) considering that we are abstracting all the theory about the gesture recognition. Also, we can access directly and easily to the raw data. Finally, the proposed prototype is described in a way that researchers can reproduce it.