Autonomous Cameramen Continued

If the robot recognizes the position of each camera actuation and compiles that data with the video (which would be aided by currently existing RealD 3D technique) you could produce a much more stable and information-rich model of what you're filming.

This would be an advantage for next-generation holographic display.
Multiple units such as these could act in synchronization in capturing a scene for 3D use.

The green and blue regions (x and y axis actuations) outline where the AI system would receive the position information from. Also note the Segway/PUMA design, for maximum stability.