Microsoft can’t hope to mount Kinect sensors at the top of every notebook PC, so it may adopt a different approach: Using the webcams that are already built into most PCs and tablets.
Michael Mott, the general manager of Xbox applications and developer relations for Microsoft, told PCWorld that the company is considering simpler alternatives, including using webcams, as an alternative to mounting a discrete sensor on top of or inside of a PC.
Last week, Microsoft revealed the latest iteration of its Kinect for Windows, slated to launch sometime this summer. Like the Kinect sensor built into Microsoft’s Xbox One game console, Kinect for Windows can recognize users and track their movements. In 2012, horror-park developer Distortions Unlimited built a motion-sensing Kinect into a 14-foot animatronic demon. More recently, Kinect has been used to let fans “suit up” with the Denver Broncos, Mott said.
Until a year or so ago, however, Microsoft essentially had the motion-sensing space to itself. Then Leap Motion launched its own sensor, and Intel moved its RealSense technology out of the lab with a formal unveiling at the Consumer Electronics Show. Now, Microsoft’s Kinect sensor looks bulky by comparison.
This summer, Microsoft will release the software development kit for the Kinect sensor, Microsoft said Wednesday, allowing developers to build apps for Kinect for Windows and sell them in the Windows Store. That, plus the new “universal apps” model that shares code between various platforms, including the Xbox, should help drive development, Mott said.
That common framework will enable new development paths, Mott said.
Intel and Microsoft are taking “two paths to the same endpoint,” Mott said. “It won’t take long for us, and I don’t think it will take us long to develop for, a different set of cameras and microphones. We’re just capturing — we have a very powerful camera, and a microphone array and you have a 100 percent delightful experience in living rooms, and in the future offices and conference rooms, classrooms.
“That’s great,” Mott added. “And you have to start from there, get people into these natural experiences with a killer user interface…. I think our path is how do we make the software rock solid, and then how do we get it to scale. If the camera you buy is not as robust as the camera that you get in Kinect for Windows, going into the future, what can I still do? I can probably still recognize you, and with a microphone array, a couple microphones, and [figure out] who’s talking and where. And that’s software-enabled, and just figuring out what the hardware configuration is.”
When asked about whether that meant using the Webcam, Mott said. “You could imagine that in the future. That’s not a plan of record. We’re really focused on getting developers to understand the fundamentals” of a natural user interface, he said.
Likewise, the Kinect could be put inside conference rooms, using the technology to “focus” on a speaker using his voice and posture. Microphone arrays could help place the speaker, but the camera could play a role, too. “If I move forward and look like I’m going to stand up, it’s basically me taking the floor,” he said.
And what might a Kinect-equipped PC look like, a decade down the road?
“I think that [PCs] will react to you, naturally, and they’ll come on, if they’re in sleep mode, and sign you in,” Mott said. “They might be able to see that you’re in a bad mood, and give you a little bit of mood music. It can allow you to be truly multitasking, so if you’re on the phone and don’t want to give away that you’re clicking the mouse and typing in the background, and move certain things across, you can.”