• The Hollywood Reporter on LinkedIn
  • Follow THR on Pinterest
AUG
8
3 MOS

'Teenage Mutant Ninja Turtles': How the CG Characters Launched New Technology (Video)

"The whole project was about performance," visual effects supervisor Pablo Helman told THR.

For VFX supervisor Pablo Helman, the key to the success of the Jonathan Liebesman-directed and Michael Bay-produced reboot of Teenage Mutant Ninja Turtles — which is projected to open in the $45 million to $50 million range this weekend in North America — was all about the performance of its four green CG martial arts experts.

"We knew that if we don't have the turtles, we don't have a movie," said Helman, a longtime supervisor at lead VFX house Industrial Light + Magic, who has been nominated for Oscars for War of the Worlds and Star Wars Episode II: Attack of the Clones.

"The performance of the turtles is the result of a combination … of the actors' performance, which is incredible and we couldn't do this without them, and in the artistry and the animators to reinterpret that performance and tell the same story that the actor told but on a different body with different proportions."

For ILM — which Bay also turned to for its Transformers robots — this challenge meant creating a new performance capture pipeline that included new techniques for body capture, which was done on location rather than on a performance capture stage, as well as facial capture.

For the latter, ILM created a new helmet carrying two HD cameras that were worn by the actors playing the turtles: Alan Ritchson (Raphael), Noel Fisher (Michelangelo), Pete Ploszek (Leonardo) and Jeremy Howard (Donatello).

This gave the visual effects team more detailed data (they captured roughly 1TB of data per day) on which to base the performance, though this was still tricky. Helman explained that while the majority of the performance develops through close-ups, some details that make up the facial performance, such as eyebrow movement or nostrils flaring, were features that were hidden behind the turtles' masks.

To get more nuanced movement in the face, new CG software was created that effectively divided a face into 200 sections, allowing the artists to focus on details in given areas, according to Helman.

A new, more complex on-set capture system was developed for the movie, allowing the filmmakers to "scientifically capture the performance on set, scientifically retarget the performance onto a digital creature, and then edit the data," Helman explained.

The ability to edit the performances was important, he added, "because this is an ensemble performance — usually four performances at a time — and it's very difficult to get four actors [to do exactly what you need] in one take. So you need to be able to edit."

For Helman, the chemistry between the actors was a plus. He related that one day on set they just started playing around and rapping. Later in the production, the filmmakers were looking for something for the turtles to do while on a long elevator ride to the top of a building. "We were looking for an honesty, and I saw their chemistry," he recalled. "I said, 'Remember this thing you did six months ago? That was so cool and so natural. Can you do something like that?' " The actors did just that, and while it wasn't in the script, it landed in the movie. (See the creation of that sequence in the above video).

Teenage Mutant Ninja Turtles isn't the first film for which ILM used performance capture on set, though this film does up its level of sophistication. The seeds go back to ILM's Oscar-winning work on Pirates of the Caribbean: Dead Man's Chest, during which it used on-set performance capture for Bill Nighy's body performance of Davy Jones and shot reference video to help to shape the facial performance of that character.

For Turtles, Helman explained that another element that makes the VFX work is that the CG characters fit realistically into the New York locations. This started with the modeling of the characters, giving them complex skin and textures.

It was also about lighting, which Helman felt was more natural because it came from the on-location photography rather than a motion capture stage. Said Helman: "I made the decision that we were going to use very little greenscreen and bluescreen. The lighting had to be right on."

Email: Carolyn.Giardina@THR.com
Twitter: @CGinLA