In the last decades, we have witnessed to the increase of the socalled Kinect-like devices, which are based on a set of low-cost sensors to acquire RGB and depth data of a scene. The high accessibility of such devices, mainly in terms of costs, has pushed their adoption as fundamental tool for gesture recognition in a large number of applications, both commercial and research-related ones. In this paper, we first discuss some of the general principles adopted by most of the main gesture recognition techniques described in literature. Then we present some application fields in which Kinect-like devices and gesture recognition algorithms have been used, ranging from educational-recreational examples to more complex and scientific fields (e.g. domotics, robotics and biomedical engineering). In two annexes, we list and shortly compare the main features of the Kinect-like devices available on the market, and we describe one of the most popular algorithm for skeletal tracking, which is the basis for the gesture recognition.
This article is authored also by Synbrain data scientists and collaborators. READ THE FULL ARTICLE