Machine-to-Machine Interaction Based on Remote 3D Arm Pointing Using Single RGBD Camera

Abstract

In this paper, we propose a prototype of machine-to-machine interaction system using a single RGBD camera. Based on the 3D arm pointing and tracking algorithm, system enables user to select multiple objects using their two arms, without infrared remote controller and wireless transmitter. A fast gesture recognition method is used to identify the instruction assignment for the target objects. The system consists of three modules, including image capturing module, 3D arm pointing and target tracking module, and interaction module. The image capture module use the RGBD camera installed behind of user to capture the arm parameters in real-world space. The pointing and tracking algorithm is referred to the environmental information to establish the master and slave objects. For interaction module, the command gesture is determined by the gesture recognition, interacting with target devices by the decoded instruction. Finally, system generates a control signal for triggering the master object to communicate with the slave object according to the fetched the instruction.