My old Adobe colleague Sylvain Paris & a team at MIT have long worked on enabling style transfer among images (see demos from 2010 & last year), and now they’re demonstrating new progress in a new SIGGRAPH paper:

Using off-the-shelf face recognition software, they first identify a portrait, in the desired style, that has characteristics similar to those of the photo to be modified. “We then find a dense correspondence — like eyes to eyes, beard to beard, skin to skin — and do this local transfer,” Shih explains.

Check out this quick demo, including the ability to apply style to video (provided it’s a headshot):

Neat stuff. [YouTube]