Scene reconstruction from multiple video streams?

I've had this idea for a while that it should be possible to use multiple videos of the same scene to reconstruct a single more detailed model of said scene. Has there been any research that's looked into that? I've tried googling but this isn't my area of expertise and don't know the correct terminology.

Examples: Modelling a room by setting up multiple cameras to capture depth, stitching together multiple handheld recordings of an event into large aspect-ratio footage, and so on.