When Is The Huntik Seson 2 Begin

Some Kinect

-tomorrow is the day here in Europe-. I must admit that I was a bit skeptical at first. I could barely test the device at the

Microsoft Spain&Portugal MVP Open Day , several weeks ago, but the more videos I see, the more convinced I am about the potential of the device. For instance, check this XBox dashboard demo:

The smoothness and precision of hand movements are amazing. It could even be an alternative to multi-touch technology. It seems that Microsoft is almost out of stock in the U.S. what means that it’s been a very successful launch. Congrats!

Adaptability and calibration

The

Kinect

includes a small electric motor that allows the device to tilt the camera up and down (up to 30 degrees), and also operates the zooming functionalities. This allows the device to properly adapt to any space. If you want to know more about the motor, you can check

this site . Sound Processing

The

Kinect

includes an array of 4 microphones, which provide:

Acoustic source localization

Ambient noise suppression
Both functionalities allow to control the XBox with your voice (see above video for a demo), and probably will enable headset-free chat on XBox Live (though I haven’t seen this in action yet).

More about Voice Recognition on Kinect

Facial and skeletal Recognition

The video camera included on

Kinect

allows to automatically recognize players just by

looking at them. So, you just need to jump in the game, and it will automatically detect you, keeping your scores, etc. This is awesome. Gesture and movement recognition

Of course, the most important feature of

Kinect

is gesture and movement reco. It can handle up to 6 players, but making motion recognition to two of them at a time. Regarding this, both hardware and software are extremely important.

Hardware was firstly developed by the company –later acquired by Microsoft- PrimeSense

. The diagram of the setup is:

As you can see, the righter part of the diagram includes a normal color camera, to detect the human visual spectrum with a resolution that seems to be: 640x480, according to some non-official sources

. It is used for facial recognition and to include images of players in games, among other things.

The other part is regarding depth, space or 3D detection, whatever you want to call it.

How can a 2D camera extract 3D information?

The secret is the IR setup it includes, with both a projector and a camera. It first projects thousands of IR dots to you and your living room, and then read that information back with the IR camera (Depth Image CMOS). Depending on the size and distance between the projected dots, it calculates the depth map. The following is a representation of the result of this depth map:

This IR dots that Kinect

projects use a similar technology to the one used in those video-recording cameras which include the so-called

Night-Shot technology . Those cameras project IR light in front of you with a small IR beamer, and then switch to a sensor which is sensible to IR light. This way, they project light ahead (what makes seeing possible), but in an invisible way to the human eye.

Knowing that, anyone else is thinking the same thing than me? I guess so…

This is what happens if you record your

Kinect

with a Night-Shot camera (I recommend you to see it in FullHD):

Et Voila! There are the dots. The IR projector seems to be projecting a matrix of 320x240 dots at 30 frames per second (again, according to non-official sources). Another example: