Add Microsoft to a list of names that includes Leap Motion, iMotion, Myo, HTC and others. That's a list of companies currently developing hand-tracking solutions for VR, and Microsoft thinks it may have the best of the bunch.
Microsoft isn't new to gesture tracking tools and many consumers will recall the Kinect, which was originally launched for the Xbox 360 about six years ago. While Microsoft had to unbundle the Kinect from the Xbox One to unburden the console from the accessory's cost, the company never quit developing gesture tracking technologies.
On Sunday, Microsoft offered the outside world an update on the progress its researchers and engineers have made with the company's Handpose hand-tracking tech.
Project Handpose is Microsoft's effort to change how people interact with computer systems, and not just virtual reality platforms. With a build that looks a bit like a bite-size Kinect, the hardware does both long and short-range hand tracking.
How do people "interact with things in the real world? Well, we pick them up, we touch them with our fingers, we manipulate them," said Jamie Shotton, a principal researcher in computer vision at Microsoft's Cambridge, UK, research lab. "We should be able to do exactly the same thing with virtual objects."
The team working on Project Handpose envisions a mainstream product that is discretely integrated into the lives of ordinary people and will allow the technology to adapt to people instead of the other way around. People can gesture hanging up a phone to end a Skype call instead of pressing a button on a keyboard, or close a window by essentially swiping at air.
The researchers and engineers are also exploring the value of haptics, the sense of touch, in hand tracking. They're looking into how to reinforce that feeling by stimulating other senses such as sight and hearing.
One of the major advances to come out of Project Handpose, as of late, is a method for personalizing the shapes of hands, doing so with as little processing power as possible. So along with being able to quickly map out a user's hands for more precise tracking, the software's agile response better enables it to correct itself whenever it loses track of its targets.
"We're getting to the point that the accuracy is such that the user can start to feel like the avatar hand is their real hand," added Shotton.
