Natal Inner Workings Interview

New Scientist has a really interesting interview with the Natal lead. It turns out that they are using a machine learning mechanism that they’ve trained with “terabytes” of sample data. It then can recognize 31 body parts within a given video frame using 10-15% of the 360’s compute capability in 10 milliseconds. It takes 160 milliseconds to identify a new user. That’s all really cool tech.

My real concern is whether or not it will be useful for anything other than Wii-style mini-games and party games. The Wii sold a lot of units, so there is some business sense there, but will the motion tracking lightning strike twice or has the fad faded?

Microsoft’s Body-sensing Button-busting controller

Here’s a video that Kotaku recently linked to that shows a little debugging info for Natal depth buffer.


~ by shaunkime on January 11, 2010.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: