|
|
 |
|
Compositing: Advanced compositing at a smaller scale Added on: Sat Dec 14 2002 |
| Page: 1 2 3 4 5 6 7 |
This tutorial once again is aimed at the more advanced audience in 3D, beginners and intermediates are welcome to try but this tutorial will primarily be aimed at the people who know Max back to front.
We are going to cover many different tricks I've picked up for compositing and intergrating live action with CG elements. I'm mainly going to try to keep the tutorials learning curve as simple as I can, but still try to get all of my points across about each technique and when to use them.
I've learnt in the past when intro-ing to a subject, although it's helpful to show off a really cool way of doing something with lots of things happening and exploding left right and centre, it does look cool but it's a lot more confusing than it has to be.
Basics of comping
First of all I assume everyone is using 3D Studio Max version 3, 4 or 5. If you're using another 3D package you're still welcome to read this tutorial since everything here applies to nearly every 3D industry package, but I will specifically using Max terms for this tutorial. When I'm talking about compositing, it usually refers to multiple images overlayed on top of eachother to create one image.
Digital compositing is a very powerful area to specialise in. Anything can be accomplished, but it does require not only a lot of talent but also a lot of thinking. Although in this tutorial I'm not refering directly at compositing such as with Flame* or Inferno* or even Digital Fusion or any of the smaller packages, I'm more refering to compositing within a 3D package.
Usually when comping CG with video ect. you use mattes (either alpha channels or masks) to layer the elements on top of eacher, the flame* or inferno* operator does this for you and actually does all of the comping, you're just the 3D monkey churning out the CG elements. But that doesn't mean you shouldn't learn compositing, it's essential since otherwise you won't be able to get this stuff out to these operators. Plus if you're working on your own animation then of course you won't have these people to do it for you :)
Below is a few images showing his this all works.
  
In the first image we have a pink teapot, this is a 3D model. When you render this image out, if you save it as a TIFF, RLA, RLF or TGA it will save the alpha channel with it. The alpha channel is the black and white image of the teapot next to it. This is an alpha channel, it's primarily an 8 bit (256 colours) image which works as a lumience mask to tell compositing packages what areas are transparent. 100% white is opac, where as black is 100% transparent, and all of the colours in between are variations of transparency.
So we want the teapot to be there but the background not since we want the teapot to sit on another background, if you look at the teapots alpha it shows the teapot's white but the backgrounds black meaning it'll be there when comped but nothing else will be. If you look above you can see the virtual frame buffer which is the window which pops up whenever you render something, and you can see the button circled is the button you use to view the alpha channel if you want to look at what it looks like.
|
 |
|
|
