When word got out that Sam Witwer and Nathalie Cox would be playing the lead roles in Star Wars: The Force Unleashed, what followed was the usual round of questions regarding what it means to star in a video game. Over the years, LucasArts has used a variety of methods to bring actors into video games. The most common has been voice performance -- where it's the actor's voice that is captured and married to the animation of his or her digital character. In the mid-1990s, we dabbled in full motion video -- remember Rebel Assault II: The Hidden Empire and Jedi Knight? -- where actors, fully dressed in costume and makeup, shot their parts in front of bluescreen or greenscreen to become the characters in digital sets during the cutscenes of the games. This time, The Force Unleashed is taking a very different approach to its lead characters, one that benefits from LucasArts' unique relationship with Industrial Light & Magic.
In previous games which featured only vocal acting performances, it really didn't matter what the actor looked like, or whether or not their faces were expressive as they spoke. Age also didn't necessarily matter, as actors would often be called to voice characters much older or younger than their real ages. In some cases, even gender didn't matter, as it wasn't too uncommon to have women cast as the voices of young boys. We had to disregard such precedent once we knew, early on, that facial motion capture (mo-cap for short) and likeness capture would imbue our lead characters with life. This approach also affected the game's script -- it's very different than those written for previous games. The Force Unleashed script is pared down considerably in terms of expository dialogue, because we're trying to convey a lot more through a look or an expression. While previous-generation games did a lot with their character animation and dialogue, this next-generation approach will achieve unprecedented subtlety of expression and nuance of performance. Now a single look can speak volumes.
The digital models of the characters have facial features made up of "blend shapes." These are the shapes that define the brow, the cheeks, the chin and the areas around the nose and eyes. By dialing in different parameters to those shapes, an animator can create a full range of emotions. The deformation of the blend shapes can be driven by two different types of data. An animator can manually input those settings, hand-crafting the facial features along a set period of time to achieve what's called key-frame animation. Or, those changes in shape can be defined by data gathered through a mo-cap session, wherein an actor has a series of dots placed along his face that can be tracked in 3-D space by special cameras. The placement of those dots is then replicated on the computer model, thus recreating the actor's facial performance on the digital model. Quite often, the final animation is a mix of both key-frame animation and mo-cap data, as animators massage the performances to achieve exactly what they're looking for.
