Sunday, December 19, 2010

First consumer machine vision product?

By now I’m sure you’ve all heard about the Kinect from Microsoft. Many of you may even be planning to gift one to the child in your life, but I suspect the offspring of many engineers are going to find this fascinating “toy” snatched away and disappearing into the basement workshop.

Why do I say this?

Because it’s a 3D machine vision product. It uses structured light and two cameras to calculate in real time the coordinates of people who stand in front of it. But don’t take my word for it, take a look at these three product teardown reports (all culled from the pages of

Inside Xbox 360's Kinect controller

Kinect's BOM roughly $56, teardown finds

Teardown: Kinect has processor after all

1 comment:

Anonymous said...

It uses only one camera for 3D data, another one is for RGB image.

"A CMOS image sensor detects reflected segments of the infrared dot pattern and maps the intensity of each segment to a corresponding distance from the sensor, with resolution of the depth dimension (z axis) down to 1 centimeter. Spatial resolution (x and y axes) is on the order of millimeters, and RGB input from a second CMOS image sensor is pixel-aligned to add color to the acquired data." from the first link.