This site may earn affiliate commissions from the links on this page. Terms of use.

movidius-machine-intelligence

A large reason the electronic devices around us are getting smarter, and hopefully more useful, is motorcar learning. By building complex models of information, then training those models, tasks as various every bit facial recognition, language translation, and autonomous driving can be accomplished. In add-on to the need for huge amounts of compute power to train those systems, they all tend to be both very power and processor intensive to run. That has kept most of them tethered to plug-in devices — like the Kinect — or requiring large batteries, like those found in a car. For instance, Nvidia's Drive PX 2 trunk-mountable car calculator will crave liquid cooling. For mobile devices that has meant a constant connectedness to the deject, with raw data sent upward, analyzed at the information center, and the results returned.

Vision-intensive tasks like object recognition are ideal for the Myriad VPUGoogle has been trying to modify this dynamic with Project Tango, a mobile device that can exercise real-fourth dimension mapping and some object tracking, while running off merely a pocket-size bombardment. To attain that, it tapped a new kind of processor, the Video Processing Unit of measurement (VPU) chip Myriad one from startup Movidius. Past moving the processor-intensive tasks associated with computer vision into a specially designed chip, Myriad increased the performance of, and decreased the power requirements for, the vision-related functions of the Tango device. Movidius claims at least a factor of 10 savings in ability, along with an 80% reduction in both infinite and cost over competing technologies — all compelling stats when it comes to mobile device design.

Beyond Project Tango: Using Movidius for mobile auto intelligence

Now, Google has broadened its relationship with Movidius, announcing that it volition be using the company'south newest and virtually powerful VPU, the Myriad M2450, to assist bring more intelligence to a wider array of mobile devices. The Myriad isn't limited to running vision-related applications, either. Google will use Movidius's software evolution environment to port its avant-garde neural computation engine to the chip, and so that a wide-variety of deep-learning-based algorithms can be run in real time.

Movidius's development board includes its chip, sensors, and a reference camera, and is instrumented for power measurementBeing able to run deep-learning-enabled tasks locally will reduce dependence on the cloud, thus reducing latency and privacy bug. For example, your phone could recognize your friends in a photo without you needing to upload it to the cloud. Remi El-Ouazzane, Movidius CEO, explains, "The claiming in embedding this technology into consumer devices boils downward to the demand for extreme power efficiency, and this is where a deep synthesis between the underlying hardware compages and the neural compute comes in."

Unfortunately, there aren't any details yet on any new Google products that will utilise the Movidius chips (and at that place was no mention of them at the Lenovo and Google Projection Tango phone annunciation), only given the importance of calculator vision and machine learning to the future of mobile devices, I'm sure we'll be hearing more before long.

Now read: Artificial neural networks are changing the world. What are they?