I briefly worked with a guy whose previous startup had been premised on the idea of combining several consumer-grade phone-cameras into a small array and using

I briefly worked with a guy whose previous startup had been premised on the idea of combining several consumer-grade phone-cameras into a small array and using software to build a far better image than any one of the cameras could accomplish individually, yet remain cost effective because these things are sold in such absurd scale. I don't know what became of it, but it sounded plausible (at least the computer vision parts that I understood were sane-sounding). Does anyone know if people are doing this sort of thing? (I'm aware that the expensive iPhones have more than one camera but IIUC that's for different reasons).