Mixed Reality Glasses: Low-Power IoT System for Digital Augmentation of Video Stream in Visual Recognition Applications

Abstract

Mixed Reality (MR) is a promising technology in the scope of Internet of Things for a number of visual recognition applications including professional training, medical applications and geodesy. However, the state-of-the-art approaches require extra facilities for the visualization and positioning of the virtual objects in the real world. In this work, we propose the approach for video stream augmentation which relies solely on the positioning of virtual objects referenced to the absolute reference frame. This approach can be applied to a dynamic platform with six dimensions of freedom. For testing purposes, we design fully autonomous MR glasses based on low-power embedded systems and demonstrate the representation of a dot, line and cube in the glasses. Experimental results demonstrate that the proposed approach is successfully realized on the experimental testbed.