In scooping a visual effects Oscar nomination, Travis Knight’s Kubo and the Two Strings became the first animated film to do so since 1993’s The Nightmare Before Christmas—a recognition due, in part, to the work of visual effects supervisor Steve Emerson, who also this month received the VES Award for Outstanding Visual Effects in an Animated Feature, alongside Knight, producer Arianne Sutner, and Brad Schiff.
With Knight’s Oregon-based LAIKA since the Coraline days, Emerson and the visual effects department are involved in the studio’s productions from the get-go, working in concert with specific artist inspirations to bring a unified aesthetic to a project. Speaking with Deadline, as LAIKA cooks up their next animated extravaganza, Emerson dissects the studio’s visual effects workflow, while giving a glimpse into the drives and innovations that make the studio unique.
You’ve worked at Laika for a number of years. How did you get started there?
I joined during the production of Coraline, and specifically came on to help lead a team that was responsible for the cosmetic work that was happening on the puppets. We dealt with a lot of the rigging removal, and we ended up absorbing some of the compositing work, as well. After Coraline finished up, I transitioned into doing some of the stage support for visual effects as we got into ParaNorman testing, and I’ve been here ever since.
I was actively working as a visual effects supervisor for The Boxtrolls when I first started seeing images that would eventually come to be realized as Kubo. I knew pretty quickly that it was going to be a daunting effort, and ultimately, we were going to be creating something that, in terms of scope and look, was far beyond anything we had previously achieved here at LAIKA.
Kubo is the first animated film to receive a visual effects Oscar nod in over two decades. How does the recognition feel?
It’s incredibly satisfying. It was important for us, as we started speaking about Kubo and putting the word out about the type of work that went into creating those images, that people started to realize that even though it’s an animated film, that our processes here are very much rooted in a live-action visual effects workflow. We have actors that we shoot on green screens. Those actors just happen to be very small, and they happen to be brought to life one frame at a time. We have sets; those sets just happen to be extremely small. We extend those sets. We add effects. We use a lot of the same software and practices that a Marvel movie would use. The difference here is that we’re doing it in a very small environment, and it takes us a heck of a long time to get those performances out of our actors.
Generally speaking, what can visual effects bring to an animated film?
The film itself features countless environments, crowds of characters. We’ve got fast-paced action sequences, giant monsters, water systems. There’s snow blizzards, there’s rainstorms. We have longhaired characters that have flowing robes. There is an underwater sequence. We have smoke monsters, we have an origami dream world, and then we have many expansive, Kurosawa-esque vistas in the film, as well. It’s just not the type of scope that you would ever expect to see in a stop-motion film. That being said, our director, Travis Knight, he wanted to make a stop-motion David Lean film.
We had to figure out how to be able to tell that story, and to deliver those visuals for him without making any type of concessions. Some of that ends up coming largely by using technology. When we talk about environments, we always try to get as much in camera as we possibly can. When it turns out that we end up having to shoot puppets on a limited set, and we have to extend environments, we work very closely with the stage team, the art department, to make sure that everything that’s being developed digitally is rooted in practical materials, and reference that’s shot out on the stages.
Beyond that, it comes down to the limitations of the stages. For instance, if you have a snow blizzard, you can’t grab ahold of an animated snowflake one frame at a time.
That becomes a challenge that ultimately ends up becoming digital, but at the same time, we still ask them for animation tests and practical visuals that we can draw characteristics from, just to keep it rooted in a stop-motion, hand-crafted look, so that ultimately, it feels like it belongs in that universe.
What kind of research process did you go through for Kubo?
The look of the film is very much based in the look of Japanese woodblock art, and specifically a woodblock printmaker whose name is Kiyoshi Saito. There’s an economy of visual information, he’s got elegantly simplified shapes, there’s a soft color palette with bright colors, and then there’s also the texture that’s created by the woodblock printing process. It’s got a wood grain texture, so a lot of the key art that we were initially inspired by felt like Kiyoshi Saito’s art. The challenge for the studio was, how do we make a live-action stop-motion film feel like woodblock art?
We keyed off of all those characteristics. If you look at Kubo, and if you pay attention to ground planes, building surfaces, costuming, even the water, everything that you see in a given frame reflects those characteristics. It came down to us trying to figure out ways of getting woodblock patterning into each and every raindrop in the storm system. Those discussions happened. [Laughs]
How are the art department and the visual effects department integrated at LAIKA?
One thing that’s unique about LAIKA is visual effects is very much involved from the beginning of a given production. We are there during pre-production, when the art is getting developed. It’ll all start with artwork which is being generated by the art department. We’ll review that with the art director and the production designer. The task at hand becomes, okay, how do we bring this piece of art to life as literally as possible?
We’ll take that artwork and we’ll head out onto the stages. And then the stage animators, the lighting and camera team, they’ll get to work putting together physical testing and shooting references out on the stages, for whatever’s reflected in that art.
It really depends on what the given image is. If it’s effects—like certainly for the water system that was developed for Kubo—that’s a great example of artwork that was translated into stage testing, which [rigging supervisor] Oliver Jones and the animation team had done. They built a contraption that was made up of an undulating grid pattern that they would cover with all these different types of materials. They would animate it out on the stages, and we looked at all the tests, and ultimately everyone started falling in love with the work that they were doing with these garbage bags, because the weight of the garbage bags would fall between the grid spaces, and it would create this kind of scoop patterning. The scoop patterning ultimately ended up being a key characteristic of the water system for Kubo.
Once those stage tests were complete, it all comes over to visual effects. We’ll have stage testing, we’ll have all the artwork that was done. And then we start to iterate, working in close collaboration with the production designer, the director, the animation team. We go version after version until we start to find something that feels like it’s reflective of Kubo’s world. It feels physical, it represents the style of the film—hopefully it’s elegant, it’s beautiful, it’s realistic, to the sense where its not distracting to the audience. But at the same time, it’s not pure realism in the least bit.
For this project, you worked with the largest stop-motion puppet ever created, standing at 16 feet tall. What kind of challenges did that create for you?
That would be the Hall of Bones skeleton, which was 16 feet tall with over a 22-foot wingspan, and weighed over 400 pounds.
There were a lot of conversations up front about whether or not that should be a CG element. One of the things that I love about LAIKA is that, when we looked at the shots that we were going to have to do, and we realized that we weren’t going to be able to pull it off with a miniature puppet, and that it was going to require us building something as mammoth as the Hall of Bones skeleton, that we didn’t shy away from it.
On the visual effects side, the biggest challenge was that the puppet itself, its arms weighed so much, that they ended up having to create this nest of cabling above it, and they would have cables that would hang down from the cabling up top in order to be able to register the arms in a given pose. The problem with that is that as the puppet would perform, all those wires would cross in front of the puppet. In the end, that’s all work that needs to be addressed in post-production.
What can you share about the 3D printing system at play at LAIKA?
The primary task for that part of the studio is dealing with facial animation, and then also innovating on the facial animation. When I go to the movies, I want to be told a great story, I want to see some amazing visuals, and I want to be moved emotionally. Typically, if I’m being moved emotionally, it’s because I’m empathizing with the characters, and I believe in those characters. The task at hand here at LAIKA is, we need to do that, but we need to do that with puppets.
There’s a push here to get as much realism as possible into those facial performances, into the body performances, into these worlds so that hopefully, a viewer, when they sit down to experience a LAIKA film, they’re just enjoying the story, they’re being drawn into it, they’re buying in, and they are believing.
For Kubo, in the case of that particular puppet, he was capable of over 48 million different expressions. Some of the expression changes were less than the width of a human hair. It’s crazy stuff. They are constantly innovating in order to bring more realism to those facial performances. They also printed up the Moon Beast—that was our first fully 3D printed puppet, and it was made up…I believe the number was over 880 individual parts, all of which were 3D printed. So they’re a huge part of the effort to make these films here.
Travis Knight has said that he is looking to truncate the time between releases, eventually putting out a movie a year. What are the major changes you’ve seen in the company throughout the years, and what are your thoughts on that idea, given where you’re at now?
It’s important to understand that if you compare a film like Coraline with Kubo, and the number of characters, the number of environments, the scope of those two films, the amount of effects, the digital work that’s being done, we have come so far, in terms of what we can achieve over the course of five years developing a film, and two years in production. We’ve come so far, and we’re doing it with just about the same resources.
Our infrastructure has become much more refined and powerful, the software has gotten better and better, so with the same resources, we’ve been able to deliver films much larger in scope. So now, when we talk about doing a film a year, I think it really comes down to just more space, more talented artists, and we’ll make that happen. When that happens, I have no idea, but I wouldn’t be surprised if within the next ten years, we’re delivering a film a year.