A survey on dynamic hand gesture recognition using kinect device

Abstract

In Human Computer Interface (HCI) technology, Hand Gestures Recognition (HGR) is a diverse field. In Dynamic Hand gesture recognition (DHGR), an unprecedented work has been done over few decades and it is still growing day by day. HGR has been extensively used in other scopes like biomedical, gaming and entertainment, research and monitoring etc. Because of its versatile utility, HGR is getting popular among the people, as it is making HCI more efficient, natural and user friendly. For the purpose of accurate segmentation and tracking a controller free and fascinating device, Kinect was introduced. In this paper Kinect based algorithms are discussed and addressed. Algorithms for DHGR are compared and particularly focused on Hidden Markov Model (HMM) and Support Vector Machine (SVM). At the end, it is observed that recognition accuracy improved significantly with Kinect device due to its good interactive features, efficiency and accuracy.