We're reaching a point at which the Kinect might be more impressive for everything it does except gaming -- and projects like this one might seal the deal. DSLR footage spread over a 3D-mapped body? Here's the future.
As the Creators Project points out, the technique of pasting human faces and frames on computer forms isn't new and was recently taken to new levels by the incredible graphics work in Rockstar's L.A. Noire. It's now hard to watch Mad Men without thinking of the L.A. Noire guy. But what if you could get similar results with the (relatively) cheap gear you already own? A crew going by RGB+D shows it's possible -- and it looks amazing.
There's an amazing amount of coding prowess behind the project, which RGB+D hopes will "break down the notion that CGI and live action are different approaches to filmmaking". But hardware is stupid simple. Slap a DSLR to a Kinect and let the feeds mix. The yield is a gorgeous, ghostly, "human" form that shimmers and dances with the Kinect data's imperfections. You have the realism of a person's face, with the sci-fi hologram look of a desperate Princess Leia or flickering AI. It's imperfect -- it's nascent. It can only get better as it's refined, reaching the sophistication of L.A. Noire -- but in the meantime it's utterly beautiful, wonderful in its own lo-fi meets high-tech ingenuity. [RGB+D via Creators Project]