In current systems, augmented reality graphics is generated at a central broadcast facility or studio where it is combined with the video that is transmitted to subscribers. By contrast, in the described system, the studio does not generate the graphics, but transmits video together with real-time metadata to the end-user set-top device. The end-user device generates the augmented reality graphics, using the metadata to determine positional and other parameters for displaying the graphics. Shifting the generation of augmented reality graphics to the consumer level facilitates end-user customization and individualized targeting of information by a broadcaster or advertiser.