Those of you who are Slashdotters have probably already seen this today, but it has got to be one of the coolest things I have seen in a long while.
http://grail.cs.washington.edu/projects/videoenhancement/videoEnhancement.htm
Using Photographs to Enhance Videos of a Static Scene
Wow, that's pretty tight. The only question I have is: what are the limitations? I mean, a pan over a static scene is one thing, but what about an actual, active, complex video?
Speaking of cool video processing algorithms, I was at SIGGRAPH last week and was blown away by this paper: