obtaining a three-dimensional model representing a geographic location, wherein the three-dimensional model is generated by a machine-learned model based on a series of two-dimensional images of the geographic location that have been previously captured by a camera following one or more paths through the geographic location;

accessing user-generated media content associated with the geographic location;

receiving path information representing at least a portion of a path through the three-dimensional model representing the geographic location, wherein the path information is generated based on a path of the camera as it was previously moved through the geographic location to capture ordered series of two-dimensional images of the location for use in generating the three-dimensional model;

selecting one or more pieces of user-generated media content based, at least in part, on the path information;

modifying the three-dimensional model representing the geographic location to include the one or more pieces of user-generated media content based on the path information and a portion of the three-dimensional model to be displayed to a user, wherein the pieces of user-generated media content are presented within one or more visual pop-outs in the three-dimensional model; and

providing the three-dimensional model representing the geographic location for display to a user.