Alita: Battle Angel is one of my most-anticipated films of the year. However, the trailers have been driving me crazy. My eyes can’t quite comprehend what they’re seeing. What’s animated and what isn’t? Everything looks real and surreal at the same time, and two or three minute bursts of short clips are too quick to comprehend.
A little research cleared up my biggest question: Alita (played by Rosa Salazar) is CG. Christoph Waltz and most of the other actors are live-action.
Weta did visual effects for the film. They are famous for their work on Avatar, the Planet of the Apes prequels, and Gollum from Lord of the Rings. Salazar wore a performance capture suit for Alita. A Weta crew member followed Salazar around with a computer attached to a camera on Salazar’s face. They made sure the dots on her face were all in the right place, and that they were collecting all the right information for her facial capture. Salazar says of the suit, “I was a piece of technology walking around.”
Much like her character.
“The way the performance capture process worked,” says producer Jon Landua, “is we had a number of cameras placed around the set that would record markers on the actor’s physical body. The system would interpret that marker and put it into three dimensional space. [And then] create a skeleton to go inside a CG model to drive that CG model’s performance.” (To see how this works, see the video below.)
Landau explains the advances Weta made in facial animation. “What they’ve really worked on is on the inside-out. Working on a system that is not driving the facial performance from the outside, but understanding what the muscles are doing under the skin, and then moving the skin…One of the artists talked to Rosa about all of the idiosyncrasies she does that we’re not supposed to be able to do, this eyebrow will go up but this lip will go down. No one else does that, but Rosa does it…It was Weta learning and teaching their system from the inside out, which is a brand new thing they’re doing and really pushing the technology.”
Eric Saindon, visual effects supervisor for Weta and two-time Oscar nominee for The Hobbit films, expounds further on this topic. “We’re now able to work at the level of the facial musculature—so it’s no longer about just moving the surface skin, but moving the underlying muscles. You can see it in how the movements of Alita’s face look so much like Rosa’s. We spent hundreds of hours just working with Alita’s mouth, because what makes even a big action scene work, is getting the most human expressions, and Rosa has a very expressive face.”
Instead of using computer graphics, a live set was constructed. “We’re in a live-action world that has digital characters brought into it,” says Joe Letteri, senior visual effects supervisor for Weta.
“Rosa’s not performing against a green screen,” says Jon Landau, “We put a 97,000 square foot set on the back lot of Troublemaker Studios, where Keean could go on a bridge in the rain with Rosa right there, and tenderly touch her cheek in a romantic scene that you couldn’t do before. But Weta pushing the technology is allowing the actors to really do performances with each other.”
Rosa Salazar confirms, “I was given the gift of…existing in a practical environment with practical props and practical people and, with the exception of one day working on a green screen stage, all the other days were in the real true environment of that set.”
One of the main sets is Iron City, a future metropolis which has gone to waste, though Alita sees it as beautiful. “Since she has no memory, she sees everything in a beautiful way,“ says director Robert Rodriguez. Creating Iron City fell to Steve Joyner and Caylah Eddleblute, production designers who have worked with Rodriguez ever since From DuskTill Dawn. “We knew going in that Robert is all about depth,“ says Joyner. “We knew he would want a city that seems to go on forever. So we knew we would have to create an intricate network of streets, alleys, and corridors that all interconnect. Ultimately, the sets were so complex there were nooks and crannies even Caylah and I didn’t know were there!”
A scanning crew would scan every environment, so they would have a 3-D model for every set they shot in.
Of course, not all of the preparation for the movie has been effects-based. Salazar started physical training for the demanding role the day after being cast. “You don’t want to cast someone and then you get two takes out of her, and then she’s exhausted,” she says.“ You want to get your endurance level up for something as physical as this. I went into training the very next morning and I trained for close to five months for a few hours every single day. I did some muay thai, some kung-fu, staff work, kick boxing.”
If you’re as excited for Alita as I am, you’ll want to stay connected with RunPee. Like us on Facebook. Follow us on Twitter. Keep an eye out for our upcoming Alita review, and get the Peetimes for Alita through the RunPee app.
Note: The quotes used in this article came from the 20th Century Fox panel at New York City Comic Con, and from the production notes.