This site may earn chapter commissions from the links on this page. Terms of apply.

Intel vision for IoT

Custom silicon vendor Movidius has attracted a lot of attention for its high-performance, low-power chips that accept powered vision applications like Google Tango, too equally making motorcar learning possible on mobile devices. At present information technology has received the ultimate compliment. Scrap giant Intel has acquired information technology to help accelerate its RealSense project and other efforts to provide computer vision and deep learning solutions. Intel is expecting to come across Movidius technology deployed in drones, robots, and VR headsets — in addition to more traditional mobile devices such as smartphones and tablets.

The Movidius advantage

Intel announces Movidius acquisitionAbility requirements are the traditional Achilles heel of mobile solutions that require substantial computation, with vision and automobile learning being two of the nearly farthermost cases. By creating optimized, custom silicon — its Myriad fleck family unit — Movidius has reduced the power needed to run car learning and vision libraries by well over an order of magnitude compared to a more-general-purpose GPU.

RealSense

Later on a lot of initial excitement, Intel's outset-generation RealSense products — designed to provide devices with a 3D view of their surroundings to support mapping, navigation, and gesture recognition — faltered due to technical shortcomings. However, Intel has more than re-doubled its efforts, and is aiming to make RealSense the eyes and ears of the Cyberspace of Things, which Intel believes will comprise over fifty billion devices past 2020. Intel Senior VP Josh Walden likens vision processors such equally Movidius's Myriad to the "visual cortex" of IoT devices.

Intel taking aim at Nvidia's GPU strategy

Intel RealSense devices can be used for both gesture and facial recognitionThis move takes Intel further into Nvidia's home turf. Nvidia has bet big on high-functioning computing for AI, cocky-driving cars, vision, and VR — the verbal markets Intel is trying to movement into with its RealSense platform, and now the Movidius conquering. This pits Nvidia's strategy of providing the most possible general calculating power per watt versus Intel's custom silicon.

On paper, the advantages of each are fairly straightforward. General purpose GPU (GPGPU) computing provides the near flexibility and adaptability, while custom silicon can exist more efficient when running a specific job or library — once information technology has been adult. In the marketplace, look to see plenty of blueprint wins for both Intel and Nvidia, and some leapfrogging of each other as subsequent production generations roll out from each.