According to various embodiments, a stream of image frames depicting a structure in a scene are obtained. The stream of image frames may comprise first image frames from a first imaging device and second image frames from a second imaging device. Using the first image frames and the second image frames, a wireframe of at least a portion of the structure is generated. From the wireframe, as-built dimensions may be identified, materials estimates may be determined, and/or data for a fabrication device may be generated, for example.