A review on gesture recognition using kinect

Abstract

Kinect provides an interesting interaction between user and device with controller-free entertainment environment. It has been used widely in various fields; research, surveillance, medical, entertainment and etc. It is necessary for user to communicate and control a device in natural and efficient way in human-robot interaction based. Hand-controllers and electromechanical device have been used by humans to control robots or machines but there were some constraints in several factors of interaction. As of that, gesture recognition techniques were introduced to overcome those problems. Three sensors in Kinect; an infrared camera, an infrared laser projector and a color camera are used to track and recognize skeletal and human body. In this paper, analysis on gesture recognition in Kinect will be discussed. Several techniques will be compared and the best gesture recognition technique in term of accuracy and efficiency will be chosen in the end of study. At the end of the study, a technique will be proposed to increase the precision of the human gesture recognition using Kinect for better performance in human-robot interaction.