Star Wars Episode II: Attack of the Clones -- Catching Up With Rob Coleman, Animation Director

Karl Cohen sits down with Rob Coleman, Episode II's animation director, and learns how the ILM team combines many different techniques into one seamless digital world.
Posted In | Magazines: AnimationWorld, VFXWorld

A Motion-Capture Finale With A Cast Of Thousands
Most of the clones in the film are moved using motion-capture. They had to be moved in a realistic way as they are supposed to be human clones inside of armor. There are shots of clones marching, walking, standing and shooting. In some shots, there are thousands of clones fighting thousands of droids. It would have been too time consuming to try to animate those sequences by hand.

With so many elements, a procedural program where the models were automatically moved in battle would surely have been a blessing. However, for Coleman there was no such short cut. "Unfortunately, we haven't reached that level yet. We did a lot of motion-capture sessions where we had people running and shooting, running and falling, jumping and shooting, and marching. We created a library of elements with all these different actions, and then a choreographing team choreographed the fights."

"I think of the big complicated shots as having three levels. We have the foreground action that most of the audience is going to see. We have the mid-ground level where you may perceive some of the action. The background action is really just there for texture. You can't see what's going on there unless you still frame it on a DVD." In battle scenes, there may be upwards of six droids or six clones in the foreground. The mid-ground may have something like 20 that are pulled from the library of elements. When they build a mid-ground scene they may have 15 or 20 stock sequences to work with. In the background are what ILM calls "sprites." They are tiny bits of action applied to low resolution clones. You will never be able to tell the difference because they are not big enough to see. So the foreground is mostly hand animation, the mid-ground procedural animation and the background is simply pulled from key frames and motion-capture action. Nobody has to retouch the artwork. If feet are sliding on the ground or arms are passing through somebody's torso, nobody will know. They are just there for background motion.

Coleman says, "We are thinking about the same kinds of things that everyone has been thinking about since people began making movies. Where do we want the audience looking? In the layout of the shot, what is the flow of the action? We have a very educated idea of where people will be looking because we are forcing them to look there. We make that section of the shot look good and then we layer in everything else. We storyboard the action just as you would on any animated film. We create animatics, so we prove to ourselves in a rough cut they're all working. And then, we layer in the final animation. We are not surprised when we see the final results."

Summing It Up
Coleman summed up his involvement with the film by saying that pre-production was getting an idea of what is coming, production is being there to witness how the original ideas get changed and post-production is seeing how it all fits together and how it changes once again. He is used to changes. Some happen when they are shooting. George Lucas has been known to restage a scene so one that was planned as a series of close-ups ends up being long and medium shots. That means more characters, additional set details and other elements may be needed to complete the scene. Coleman is used to seeing the environment change from the pre-production paintings to what George wants when they are filming. "It might be more characters, more windows, more architecture. In terms of characters, costumes can change from the drawing board to the final execution because George may make an aesthetic decision to change the look of a digital character."









Comments


xerrQgH (not verified) | Mon, 08/29/2011 - 04:02 | Permalink
It is a well known fact that you must always make an ovious story excuse for mixing a CG chacter with live ones, just as you do for cell chacters with live ones as the audience can never ever be fooled into thinking the cg was photographed. CG by it's very nature, can never be photorealistic enough to pose as a photographed phisical thing. so what story excuse was offered for the transformation of yoda. was he said to have been holographing, or in anoter dimension? what about the other cg?
Matt Darby (not verified) | Fri, 06/28/2002 - 00:00 | Permalink

Post new comment

The content of this field is kept private and will not be shown publicly.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Allowed HTML tags: <a> <em> <strong> <cite> <code> <ul> <ol> <li> <dl> <dt> <dd>
  • Lines and paragraphs break automatically.
  • Use <!--pagebreak--> to create page breaks.

More information about formatting options

CAPTCHA
This question is for testing whether you are a human visitor and to prevent automated spam submissions.