Visualizing particle velocity from dual-camera mixed reality video images using 3D particle tracking velocimetry

Abstract

Mixed reality (MR) systems integrate diverse sensors, allowing users to visualize and interact with their surroundings. Mixed reality headsets typically include synchronized front-facing cameras that, among other things, can be used to track tracer particles (such as snowflakes) to estimate particle velocity field in real time. The current work presents a 3D particle tracking velocimetry method for use with MR devices, which combines binocular disparity and various monocular cues to estimate particle distance from an observer. This distance information is then incorporated into a particle tracking velocimetry algorithm to generate a three-dimensional visualization of the particle velocities. The resulting mixed reality particle tracking velocimetry (MR-PTV) approach was initially tested using synthetic particle data obtained by discrete element method simulations, resulting in a detailed error assessment of the method. The approach was then experimentally validated for particles transported in a wind tunnel and in a water flume flow using the Microsoft HoloLens 2 MR headset to image the particle motion. The resulting MR-PTV system can be used for mixed reality particle velocity visualization in a variety of industrial, scientific, and recreational purposes.

Graphical abstract