The Kinect runtime processes depth data to identify up to six human figures in a segmentation map. The segmentation map is a bitmap with pixel values corresponding to the index of the person in the field-of-view who is closest to the camera at that pixel position. Player segmentation data is only available in the depth stream when skeletal tracking is enabled.

The player segmentation data is also commonly referred to as player index data. Although the player segmentation data is a separate logical stream, the depth data and the segmentation data are merged into a single pixel value for each frame. The value "0" indicates that no person was found at that location, the values "1" through "6" identify players. The values "1" through "6" map to elements 0 through 5 in the array of skeleton data.

One common use of the segmentation data is as a mask to isolate a specific user or region of interest from the color and depth images.

http://msdn.microsoft.com/en-us/library/jj131028.aspx

搞定