Microsoft’s next version of the HoloLens may be better at browsing through reality than the current version of the mixed-reality headset – all credits to a new coprocessor the company announced.
This second version of HoloLens’ custom multiprocessor, the “holographic processing unit,” (HPU)integrates artificial intelligence into the HPU. This will allow HoloLens to perform the deep learning, which is usually done in Cloud.
HoloLens is designed to be a complete holographic computer, with the sensors, battery, and display integrated in the headset. This enables the HPU experience reality without latency. Latency can be a real deal-breaker in these mixed-reality applications.
Accommodating an AI coprocessor into HoloLens is the technology you would be looking out for if you want to design intelligent mixed reality devices. Both mixed reality and artificial intelligence is the future of computing. And HoloLens will benefit from the AI silicon. Better accuracy in head tracking and geo-registering content — sealing digital content to the real world is the main benefit, with the offshoot of a platform gradually expanding in capability.
Additionally, silicon might enhance performance and user safety also. With the help of a high-performing AI chip, there will be lesser need for processing, resulting in a faster, more secure and private HoloLens. This speed could in turn have another benefit to the entire HoloLens landscape. It could lead the way for entirely new kinds and categories of augmented reality solutions for consumers and businesses.
However, Microsoft isn’t the only company which thinks that more processing is required at the endpoints of a network – especially in the wake of the growth of the Internet of Things.