Intel is developing depth-sensing 3D camera technology that could go far beyond detecting mere shapes and motion to mapping the contours of physical objects and even the emotional state of individuals in its field of vision, according to a report from IDG News.
Intel director of perceptual products and solutions Anil Nanduri told IDG that this technology would “bridge the gap between the real and virtual world” through creating a level of interaction with our devices that goes far beyond the keyboard, mouse and touchscreen interfaces.
“You’ll add the ability to sense your excitement, emotion — whether you are happy or smiling. The algorithms and technologies are there, but they are getting more refined, and as they get more robust, you’ll see them,” Nanduri told IDG.
For instance, the camera sensor can extrapolate whether you’re enjoying a particular video game by tracking changes in your facial expression. It can determine whether a child engaging in a reading exercise is having difficulty with a particular word by detecting where his or her eyes are focused on line of text. It could let you manipulate items on screen just as you would in the physical world – using a virtual hand to pick things up, set them down and turn them on multiple axes.
Refined motion detection technologies certainly aren’t new. Microsoft’s Kinect has been in the market for years, while startups like Leap Motion are pioneering new motion-based interfaces for other consumer devices. But Intel seems intent not just on capturing motion but detecting specific objects or traits and extrapolating what the movement of those objects or features mean.
Nanduri told IDG that the sensor could trace the contours and physical characteristics of an object and its field of vision and use that info to identify what those objects are. That could become a useful a tool in 3D printing, acting as an inexpensive 3D scanner. And though Intel didn’t cite this example to IDG, such object interpretation could be a key component in connected car and autonomous driving systems being developed by Google and the automakers. Sensors within the car could not only detect the objects around them, but also could identify what those objects are – whether car, truck, deer, person or construction cone – and what their potential behaviors might be.
Intel expects that this depth-sensing technology will be available in the next few quarters, first in standalone webcams and then embedded into laptops and ultrabooks in the second half of 2014. Eventually the technology would make its way into smartphones and tablets. We’ll be talking about new types of UI and experience design at our RoadMap conference in November in San Francisco.
Though Intel will probably tread carefully when positioning the technology. This isn’t the first time it’s talked about embedding intelligent cameras into devices. Intel’s planned TV service, OnCue, was supposed to come with a camera sensor that could identify specific viewers and tailor content recommendations and even ads to their historical tastes. Intel, though, looks to have dropped that feature from OnCue’s launch, probably because of pushback from consumers over privacy.
Feature image courtesy of Shutterstock user Yes Man