Researchers develop a 360-degree holographic display {Engadget}

Sep 3rd 2007 11:12AM Well - you could just use X amount of cameras to capture a person live and in 3D. With todays CPU power a lot less cameras would be needed since it's possible to calculate a LOT from just to cameras positioned right. So - maybe 4-8 cameras are needed at the bottom, and the same at top.

Also - making the picture bigger has at least 2 things to it. To get X + Y bigger you could just add more mirrors and link these - but getting the Z axis "taller" might be more difficult seeing the picture never goes out of the mirror boundaries :)