Last month, it was reported that Apple is working on a “revolutionary” front-facing camera system for the long-rumored iPhone 8, and now the Cupertino company has been awarded a patent for a method that’s capable of detecting faces in a video feed using depth maps when people are located at different distances from the lens.
As outlined in the documentation published by the US Patent and Trademark Office (USPTO) and reported by MacRumors, the system applies depth information to existing face detection algorithms and scales face window sizes according to their depth coordinates — so the further away a face is from the lens, the smaller the capture frame around it.
To create the depth map required, Apple is using technology developed PrimeSense, which it acquired back in 2013. The patent coincides with KGI Securities analyst Ming-Chi Kuo’s prediction that the iPhone 8 will sport fully-featured 3D sensing capabilities powered by the Israeli firm’s algorithms.
Two more modules are required, though
It’s worth noting, however, that this specific method is able to recognize faces in a digital video, but doesn’t have the ability to differentiate between different facial features, which means it isn’t a facial recognition solution by itself. When coupled with two other modules, as forecasted by Kuo, it could be capable of that, however.