A patent filed by Microsoft in July 2017 could lead to a reduction in the manufacturing costs of the HoloLens by integrating two components into one "optical system". The two components in question are the CG super imposer and depth sensor which also handles gesture recognition.
In the patent summary, Microsoft describes an optical system which allows a micro electro mechanical system to generate electromagnetic radiation with two spectral bandwidths. The first spectral bandwidth would handle the generation of the CG images which are superimposed onto the lens" as holograms. The second spectral bandwidth would scan the field of view "utlizing a terrain-mapping protocol"
The patent posits that the integration of these two components into a single optical system would eliminate the need for separate dedicated components for image generation and terrain-mapping.
Microsoft is expected to release the next iteration of the HoloLens late in the second quarter of this year, with an improved Kinect sensor, a Snapdragon 850 which will allow it to function as an always-connected PC with a 4G LTE modem and a custom AI coprocessor.
Source: World Intellectual Property Organization via MSPoweruser