Designing an empirical framework to measure the level of interest of Human

Human-robot interaction (HRI) is an interdisciplinary research field aimed at improving the interaction between human beings and robots. For effective and prolonged interaction, both interacting partners need to be identified the focus of attention of each other. In other words, need to tracks or determine the each other interest and their level of interest. Interaction between a human and a robot may influence by the proximity and angle between them. In this paper, we proposed a computer vision framework for robots that will determine the level of interest of its human partner in two-party setting. We have used the distance and angle parameters to determine the level of interest in several categories: high, medium, low, and so on. The level of interest information can be used as an indicator of people´s willingness to engage in an interaction with a robot. The proposed framework uses depth and visual information, provided by a Kinect sensor. Experimental results show that the system is functioning quite well to measure the level of interest of human with a reasonable accuracy.