


So I finally got around to figuring out how to use the depth and video feeds from my Kinect using Microsoft's Kinect SDK, and this is what I produced after playing around with some visualization for the depth data. First two images are simply the distance for each pixel multiplied by either 3, 7 or 1 (green, blue, red) and then taken modulo 256, which makes for some pretty cool images. The last one looks a little less spectacular, but is actually more interesting: I wrote a little algorithm to color chunks of pixels of arbitrary size based on the average distance within the block, and then I only apply this coloring to the pixels that are actually a person, while other objects and the background are simply individually colored according to distance.
I'll try and get around to actually producing something useful with this soon, but I thought this stuff was kind of fun.
By the way, this is all written in C# and also uses WPF.
No comments:
Post a Comment