Slashdot Mirror
Hacked iRobot Uses XBox Kinect To See World
kkleiner writes "A student at MIT's Personal Robotics Group is going to put Microsoft's Kinect to a good use: controlling robots. Philipp Robbel has hacked together the Kinect 3D sensor with an iRobot Create platform and assembled a battery-powered bot that can see its environment and obey your gestured commands. Tentatively named KinectBot, Robbel's creation can generate some beautifully detailed 3D maps of its surroundings and wirelessly send them to a host computer. KinectBot can also detect nearby humans and track their movements to understand where they want it to go."
In related but less agreeable news, "Dennis Durkin, who is both COO and CFO for Microsoft's Xbox group, told investors this week that Kinect can also be used by advertisers to see how many people are in a room when an ad is on screen, and to custom-tailor content based on the people it recognizes."
Google BotView. Little robots roaming the world making 3d models of everything.
If it rhymes it must be true.
Kinect can also be used by advertisers to see how many people are in a room when an ad is on screen,
That could be bad for those who are getting TV ad money.
When advertisers can actually measure the number of people walking out and ignoring the ads, they often start paying less for ads :).
The first thing I thought about was cheap motion capture / do-it-yourself BVH file generation; I'm a semi-pro animator & cgi guy, and this is sort of a holy grail for the basement computer graphics community.
I'm pretty sure all a person would need is 2 or more Kinects and some relatively simple code to make something that could compete with systems that cost around $5000. I waste a LOT of money on various software packages, but 5k is pretty much out of the question; an additional Xbox 360 and 2 Kinects, though... There would be a LOT more amateur and low dollar animations made.
But, after that, yeah, the level of monitoring people would be potentially opening themselves to is pretty amazing, also.
Why, yes, I AM a Pagan Libertarian.
While the work the MIT student did is noteworthy, it's really quite trivial thanks to ROS. I do robotics research using ROS, and SLAM, navigation, planning, etc. are all handled by ROS automatically as long as you provide the appropriate data streams. It's really as simple as plugging in a device. Even the gesture recognition is handled by the kinect driver and issuing commands from gestures is trivial at that point.
I think the real recognition should be given to the group at CCNY (no I don't got school there) who did the work of getting the kinect driver working in ROS in the first place, and aren't even mentioned in this article.