Google's Clips camera might be a little creepy, but in addition, it seems pretty helpful for the perfect user; deploying machine learning how to automatically snap the very best photos of your kids and pets. Nevertheless, the secret to that performance is not only Google's AI art, but additionally, it requires a specialized chip built from the Intel-owned chipmaker Movidius.
The processor in question is that the Myriad 2, which Movidius refers to as a visual processing unit or VPU. (That's as opposed to a graphics processing unit, GPU; or central processing unit, CPU.) The Myriad two is a chip tailor-made to handle machine vision tasks such as object recognition, and Movidius asserts it's the industry's first always-on vision processor. It has previously established up in Google's Project Tango apparatus in addition to DJI's autonomous drones, also helps to make their on-board vision processing more efficient.
Google continues to be thinking about Movidius' chips. As well as using their VPUs to power Project Tango, the search giant embarked on a partnership with Movidius last year to improve how image recognition works on devices like smartphones. With the launch of the Clips camera, we have a perfect example of the sorts of benefits these collaborations bring.
Clips does it all its AI processing on-device rather than relying on a connection to the cloud to scan images for familiar faces. That's good for user privacy (there's no chance of data being snaffled in-transit), but also increases the device's battery life (because it doesn't have to maintain an internet connection at all times). These benefits are the direct result of using a specialized chip like Movidius' VPU.
A growing number of businesses are turning to tech for this to enhance on-device AI. Only last month Apple introduced new iPhones complete with dedicated AI "neural engine" chips, and Huawei showed off similar ability with its recent Kirin 970 chipset. On-device AI is the future and technical silicon is helping deliver it.