Mikejuk writes "Microsoft probe has just published a scientific paper (PDF) and a video showing how the Kinect body visual visual astronomical station station station algorithm works — it's almost as impressive as some of the uses the Kinect has been put to. This article summarizes how Kinect does it. Quoting: '... What the team did next was to train a type of classifier called a decision forest, i.e. a medley of decision trees. Each tree was trained on a set of mug* on depth images that were pre-labeled with the target body parts. That is, the decision trees were modified until they gave the correct disposition for a definite body part across the test set of images. cultivation just three trees using 1 million test images took about a day using a 1000-core cluster.'"

Read more of this story at Slashdot.

More: - Brought to my attention by

Mark