Take a look at this screencap from this year's WWDC session 501 "Advances in iOS photography".
As an app developer, you can tell the camera subsystem to either use the 'video' pixel format or the 'full' pixel format.
I don't get why someone would want to choose the 'video' formats. If you're shooting a movie, chances are that you'll do post-processing anyway, so why not just take the full color range every time and post-process away the pixels you don't need?
I apologize if this question comes across as ignorant, but in an age where storage is cheap and lossless media (like music and photos) is ubiquitous, I don't understand why this is necessary.