Movidius Myriad X
On August 28, Intel introduced its new Movidius Myriad X vision processing unit (VPU), the world’s first VPU with a dedicated Neural Compute Engine. The Movidius Myriad X was developed to deliver artificial intelligence (AI) capabilities in a range of products including drones, robotics, smart cameras and virtual reality devices with both high performance and low power demands. The upshot of that is a range of autonomous AI-empowered devices that can interact with and respond to the world around them.
VPUs are a kind of microprocessor designed to accelerate tasks related to machine vision. The Myriad X is the first system-on-chip (SOC) in the world with the dedicated Neural Compute Engine, an on-chip hardware block for accelerating deep learning inferences and running deep neural networks. It allows devices to see and gain sensory input from what’s around them, understand and analyze that data, and respond to their environments in real time. With the help of the Neural Compute Engine, the Myriad X can complete 1 TOPS (Trillion operations per second) of compute performance on deep neural network inferences — all within a watt of power, and an order of magnitude faster than its predecessor.
Computer Vision, Deep Learning
Myriad X will mean that devices have faster, more powerful intelligence inside them; they won’t need to be online, for example, to interact with their surroundings or function as they’re supposed to. Some of the applications of this technology will include autonomous drones for deliveries, surveillance, or even personal use. Everything from personal robots with both emotional and situational awareness and a range of devices with contextual understanding, to security cameras and baby monitors.
“We’re on the cusp of computer vision and deep learning becoming standard requirements for the billions of devices surrounding us every day,” Remi El-Ouazzane, vice president and general manager of Movidius, Intel New Technology Group, said in a statement. “Enabling devices with humanlike visual intelligence represents the next leap forward in computing. With Myriad X, we are redefining what a VPU means when it comes to delivering as much AI and vision compute power possible, all within the unique energy and thermal constraints of modern untethered devices.”
Intel plans to continue to increase edge compute performance significantly without sacrificing power consumption. This line of new technology should enable new autonomous devices that make life easier for everyone, and enable humans to do more. El-Ouazzane, quoting Saint-Exupéry, said in a statement, “’As for the future, your task is not to foresee it, but to enable it.’”