Intel’s acquisition of Movidius is enabling the chip giant to bring artificial intelligence to a ‘Myriad’ of devices at the edge.
Intel-owned Irish tech firm Movidius has revealed a new vision processing unit (VPU) that brings artificial intelligence (AI) to drones, robots, smart cameras and virtual reality (VR) devices.
Entitled Myriad X, Intel claims it is the world’s first system on chip (SOC) that is capable of delivering a dedicated neural compute engine for enabling deep learning at the edge, especially on machines such as drones and robots.
‘Enabling devices with humanlike visual intelligence represents the next leap forward in computing’
– REMI EL-OUAZZANE
Myriad X is an on-chip hardware block specifically designed to run deep neural networks at high speed and low power without compromising accuracy, enabling devices to see, understand and respond to their environments in real time.
The chipset is capable of a minimum 1trn total operations per second (TOPS) and a maximum of 4trn TOPS, a breakthrough for deep neural computing at the edge.
Movidius: The Irish tech firm at the edge of tomorrow
Intel acquired Movidius almost a year ago for an undisclosed sum, rumoured to be in the region of $300m.
Movidius, founded by David Moloney and Sean Mitchell a decade ago, was set up to bring deep learning and AI to devices such as smartphones. It struck lucrative deals with Google to power its VR headsets, as well as with Chinese drone maker DJI to make the world’s most intelligent drones.
In July, Intel and Movidius revealed a new $79 Neural Compute Stick, described by Intel as the world’s first USB-based deep learning inference kit and self-contained AI accelerator.
“We’re on the cusp of computer vision and deep learning becoming standard requirements for the billions of devices surrounding us every day,” said Remi El-Ouazzane, vice-president and general manager of Movidius, Intel New Technology Group.
“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.”
The chip is capable of running multiple imaging and vision applications at once and can connect up to eight high-resolution RGB cameras, supporting up to 700m pixels per second of image signal processing.
The chipset can utilise more than 20 hardware accelerators to perform such tasks as optical flow and stereo depth without the need for additional computing.
The tiny SOC also comes with 2.5MB of homogeneous on-chip memory, allowing for up to 450Gbps of internal bandwidth.