I hate blogs where they post a single video clip and expect the viewer to just watch with little to no idea of what they’re about to watch. You’re just expected to give up a few minutes of your life by trusting that you have the same sense of humor or intrigue as the offer. I’m going to be better than 99% of the blogs out there and going to describe for you what you’ll be watching if you decide to click play on the video below.
A research lab in Germany has developed a software engine that, after taking 200 3D scans of human faces, can scan a single 2D picture or photograph, and will generate a very realistic 3D model which can be used to synthesize the movement of the face for speech or expression changes. It is remarkably realistic, and I can’t wait to see it’s application in movies and other moving mediums.
UPDATE (August 10, 2007): There have been advancements made to his software collection. Read about them here.
Truth be told, when I was approached by Michael Mancuso a.k.a Esselfortium to review his upcoming CD “A Terrible Flood”, with his description of it being “electronic; some downtempo, some drum’n'bass, some IDMish stuff, and some things in between”, I was apprehensive — no, not apprehensive — I was a bit nervous that I was going to be subjected to 60 minutes of the same FruityLoops drum line, with one or two synth lines layered over top; you know, the usual whenever someone on the internet produces “IDM”. Call it prejudiced, but I’ve spoken with probably 20 internet-based “IDM producers” and my God, it was all the exact same: they created one little line in some audio production program, called it a masterpiece, and decided to unleash the horrors upon my ears. Oh, and all this while refusing to take any criticisms about their so-called “art.”