So a dude called "PlayerDuce" on YouTube recently uploaded this video which shows him using the Oculus Rift in conjunction with the Kinect. He uses the Rift's sensor for head rotation, while his Kinect generates a 3D point cloud environment around him. Watch:
I haven't quite wrapped my head around how he does this, but it seems to create a 3D world (abstract though it may be) in real time, displayed around the user. Mr. PlayerDuce sees some interesting applications to this tech:
"The first thing I want to do is integrate it into a video games and demos," he tells me. "There are quite a few games and demos that don't look realistic to me in the Rift, so this will provide a reference point...Another application is telecommunication, although people may prefer 3d avatars over themselves wearing an Oculus Rift." That is likely true in some cases, though I bet you could generate and merge such an avatar into the scene. In any case, I definitely see cool things coming out of this hack.
Hat tip: Oculus' Reddit group, which has interesting comments on this demo.
Please share this post with people you like:
Tweet
I see what's going on here! Rather than having a Kinect sitting on a TV observing him move around, he mounted the Kinect to his headset, effectively adding a Depth Camera to the Oculus Rift, allowing it to recreate the world around him in 3D and display it. Very nice.
I'm not sure if the consumer version of the Rift will support positional tracking, but if it doesn't, something like this could certainly add the functionality.
Posted by: Adeon Writer | Tuesday, April 30, 2013 at 02:30 PM
No wait, what he's doing is more complicated than that. He's getting the positional data via kinect and rendering the view from his perspective. That's even more impressive.
Posted by: Adeon Writer | Tuesday, April 30, 2013 at 02:40 PM