Our future is increasingly data-driven. Yours. Mine. Everybody’s. In fact, there’s a silent revolution underway that is leveraging advances in big data and artificial intelligence (AI) to change our world in amazing ways.
Speaking at the recent Web Summit technology conference in Lisbon, Intel CEO Brian Krzanich showed how the real-time analysis, collection and delivery of data is enabling some remarkable and surprising solutions.
Take Australia’s Little Ripper Lifesaver UAV, a drone that uses an Intel® Movidius™ Neural Compute Stick to monitor the New South Wales coastline for sharks. It’s a proof-of-concept system, showing how modern AI processing can take place ‘on device’ (rather than in the cloud), leading to faster danger detection and shorter response times.
The ultimate aim is to create an algorithm that can scan live video to spot sharks in real-time.
A fleet of self-driving cars
Visual data collection and analysis is also a crucial component of tomorrow’s self-driving cars, which is why Intel and Mobileye are building and testing a fleet of autonomous vehicles designed to the limits of real-time image processing, intelligent navigation and mapping.
“We’re building a system that also collects data to allow real-time mapping,” Krzanich told the Web Summit audience. “So that, as the cars are driving along and they notice accidents, they notice construction, they’re feeding that data back and allowing the maps to be updated in real-time…. Thus allowing the cars that are behind, to have a much more intelligent capability.”
AI can also be applied to entertainment. In the US, Intel True VR camera pods can already capture Major League Baseball matches from every angle and a new partnership with Turner Sports will bring immersive VR viewing to NBA matchups in 2018.
“With Intel True VR camera pods, we can dive into the middle of the action,” said Krzanich. “When adding AI to the experience, we will be able to watch multiple games and receive notifications when something key is about to happen so we don’t miss the action.”
Of course, artificial intelligence also has some more serious uses. The ChildFinder application, developed by Thorn, uses AI and a data-driven approach to locate and identify the faces of missing children from CCTV camera footage.
“The AI system is trained to recognise when a pair of images represents the same child, regardless of the child’s age at the time each photo was taken,” explained Krzanich. “It even allows for differences in clothing, hair style, makeup and background when matching images. With the help of AI, a team can quickly scan thousands of images and identify the most likely matches, turning an almost impossible task into a much simpler one.”
Teaching machines to understand
Giving machines the ability to ‘see’ isn’t the hard part in all this. Getting them to understand what they see — to recognise patterns, draw conclusions and deliver actionable insight — is the challenge that only AI can hope to overcome.
The amount of data we generate is already huge. The average Internet user creates 1.5GB every day. With the rise of social media, the introduction of connected cars, video monitoring systems, smart factories and cloud-based entertainment platforms, we will soon be overwhelmed by a flood of information.
In some respects, we already are. How do we make sense of it all?
Data is becoming one of the most valuable commodities on the planet. It’s been described as ‘the new oil’, and just as oil transformed the 20th century, data is poised to transform the 21st. AI will make sense of it, leading-edge computing power will drive it. Find out more about how Intel can provide both here.