I don't normally do this but there is some amazing work being done at the University of Washington.
Using Photographs to Enhance Videos of a Static Scene from pro on Vimeo.
Spotted on CrunchGear.
I was impressed by smile and blink detection in my little point n shoot. That's nothing compared to what these folks are doing.
They take some still images of a scene and then a video. The still images, including HDR images, can be used to enhance the quality of the video scene. There is also a brilliant image stabalisation system that works after shooting.
My only question is just how much of this requires operator skill and how much is automatic?