as implemented by a model generation and display system comprising one or more hardware processors,

generating an exterior mesh based on an exterior of a building, wherein the exterior mesh is associated with exterior mesh positioning data;

generating an interior mesh based on an interior of the building, wherein the interior mesh is associated with interior mesh positioning data;

orienting the exterior mesh relative to the interior mesh by referencing the exterior mesh positioning data and the interior mesh positioning data;

generating a combined mesh from the oriented exterior mesh and the interior mesh;

rendering a three-dimensional (3D) model of the building from the combined mesh;

dividing the 3D model into a plurality of locations based in part on structural features identified by machine learning in the combined mesh of the 3D model;

generating first metadata that associates an image of the interior of the building with a first location of the plurality of locations within an interior of the 3D model, wherein the location within the 3D model corresponds with a geographic location identified in second metadata of the image;

determining, using machine learning, that the image depicts a first type of room in the building;

classifying the location within the 3D model as being the first type of room; and

causing a monitor array to display the 3D model, wherein user input received via a sensor embedded in a table separate from the monitor array that detects movement controls a view of the 3D model displayed by the monitor array.