Kinect doesn't really have much in common with the eyetoy. The eyetoy only captured your silhouette, and let the stuff on screen react to being touched by the silhouette. Kinect, on the other hand, uses infrared technology to track your movement in three dimensions and transfer them to your on-screen avatar.
The eyetoy just identified an object in the path of the camera, and allowed it to interact with on-screen objects. The Kinect, on the other hand, identifies humans, and can even keep track of two separate humans. In other words: The Kinect knows it is filming people, and not, say, a branch wawing in the wind in front of the camera.
Edit: The only thing they actually have in common, is that a camera is involved.