Okay so we're about five minutes into this project and we've already painted ourselves into a corner, welcome to IMM. Myself, Chris, and Sulli have decided to create a short vector animtion film that will use live action footage as the basis and then later translate that into animated vector graphics. For those of you unfamilar with vector graphics allow me to explain, or better yet let me just copy and paste something from Wikipedia:
Vector graphics (also called geometric modeling or object-oriented graphics) is the use of geometrical primitives such as points, lines, curves, and polygons, which are all based upon mathematical equations to represent images in computer graphics. It is used by contrast to the term raster graphics, which is the representation of images as a collection of pixels (dots).
So that's that, now that we're all caught up let me try to explain the many problems this creates when trying to a live human being and turn them into a vector shape. First of all from what I understand most film/cameras/film editing software uses pixels to render an image, pixels are tiny blocks of color that make up the overall image. Most of the time you can't really tell the difference between a pixel shape and vector shape, except when you begin to manipulate it. A pixel shape when distorted will show small jagged edges around curves and bends in the object, these jagged edges are the pixels. This happens you're trying to make a rounded or curved shape out of an image that has been rendered with blocks. Now when you take a vector shape and distort it you do not have that problem because the vector shape is made up of all that fun math lingo up top.
At this point you may be thinking how does this apply to this project, well shut up for a second and I'll tell you. We have to find a away to take grainy pixelated footage and change it into nice, smooth, fluid animation. To do this we have to break down each frame of the film into separete image files, place them into program that will uses vector graphics and then recomposite them, then edit it into a finished product. The problem here is that we need a program that can automate the vector animation process, and the only one that seems capable is Adobe After Effects, which neither of us currently possesses or wants to spend the money to do so. So the only other choice is to to do each frame by hand using the Live Trace function in Adobe Illustrator. However let's say you have about a minute or so of footage, well when you split that up into individual images you get somewhere around 1078 individual image files. Now let's suppose we wanted to do a five minute piece, well that hurts my head just thinking about it. And here in lies the problem, however there is a small amount of hope as I have found a way that might allow us to automate the process in Illustrator, now we've just got to test it.
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment