Computer Science student Andrew Maimone created the 3D hack for his paper (overlooked by Professor Henry Fuchs) by setting up four Kinect sensors to capture the same images from different angles. Then, using a particular algorithm and GPU filters (for hole-filling, rendering etc), the images can be merged into one 3D image. For video, the cameras go through a number of steps, such as hole-filling, coloring, and quality before meshing the for the final 3D overlay.
Although the hack still looks a little rough, it certainly paves the way for a whole different aspect of creating cool Kinect hacks without the need for adding reference points. Check out the video below to see how the 3D Kinect looks, and keep checking back on Andrew’s hack’s page to see his full paper soon.
[UNC Chapel Hill via Hack A Day]
Like this? You might also enjoy…
- DIY Inkjet Printer Prints On Almost Any Surface
- Hack Gets MeeGo Onto Nook Color, Runs Seriously Well
- Giant Earth Globe Uses 10,000 OLEDs, Looks Awesome
Get your GeekTech on: Twitter – Facebook – RSS | Tip us off