A streaming visualization gives a you a sense of data flow, order and temporal pattern.
This excludes anything that is polling for data and simply showing you the latest values.
A streaming visualization will visualize the stream of data. show the data that is flowing, where it goes and what it does to stats.
The famous video by Hans Rosling on TED is a streaming data visualization. It takes advantage of movement to give you an understanding of where data values were, not just where they are now:
https://www.ted.com/talks/hans_rosling_shows_the_best_stats_you_ve_ever_seen
This can sometimes overlap with real time data, but it doesn't have to.
Real time data is just data trying to stay caught up to now. The presentation of it is often a constant refresh of current values. A streaming visualization can be real time by streaming the latest data, but doesn't have to be. Like in the Hans Rosling example, it can be a compressed and/or selected timeframe. Even though it's in the past it can help to see the movement of the data in a streaming visualization fashion.
