A system for monitoring an industrial system may have a computing device that generates a first three-dimensional (3D) visualization based on 3D video associated with an industrial system. The first 3D visualization may have one or more 3D device visualizations of industrial automation devices, and each 3D device visualization of the industrial automation devices that may depict a 3D object that corresponds to an industrial automation device in the industrial system. The computing device may display the first 3D visualization via an electronic display and detect a gesture in a first set of image data comprising the surrounding of the user and the first 3D visualization. The gesture is indicative of a modification associated with the first 3D visualization. The computing device may receive additional 3D video associated with the industrial system, generate a second 3D visualization based on the modification and the additional 3D video, and display the second 3D visualization via the electronic display.