Putting the Kinect in perspective

Over the dozen or so years that I’ve been involved with machine vision I’ve been lucky enough to work with some of the pioneers of this industry. These guys can all tell entertaining stories about struggling with analog cameras and writing Assembler-level code, but that’s not what I value most about them. What they learnt through their trials and tribulations were two things: the overarching importance of creating image contrast (because they had no fancy pre-packaged algorithms they could bring to bear,) and how to code for efficiency (because I’m convinced that abundant memory and processor power have made us fat and lazy.)

I was reminded of this by the “A Pirate Looks at 60” blog authored by Mr. Mickey Cheatham. Mickey seems to have led an interesting life, but what really grabbed my attention was his post, “Machine Vision and Microsoft’s New Kinect.” In this somewhat lengthy article Mickey reflects on just how far we’ve come in the last thirty years, and to make his point, gives us a dissection of the Kinect. He seems to be of the opinion that this product might turn out to be more of a game-changer than Microsoft realize, and wraps up with a few thoughts on where it might take us.

If anyone needs convincing that our elders have a lot to teach us, show them Mickey’s blog post. Hopefully, it will open their eyes.