Picture this: I meet Steven Schardt, the director and creator of this wildly ambitious project called Polae, and it’s like we’ve known each other forever. The guy is the most generous American French-speaking artist you could ever hope to meet. It’s as if fate brought us together to work on this innovative octopus of intertwined stories and gameplay.
Now, when Steven came and stayed with us at the studio, we dove headfirst into developing the 3D immersive elements. It was during that time that we discovered just how delightfully insane this project was and how Steven had become a self-taught real-time engine artist. I mean, who does that, right?
Steven is no stranger to the film industry, having produced numerous films that have competed at renowned international festivals like Sundance, Cannes, and Tribeca. But when it came to sizing the scope of Polae’s first development phase, we faced some challenges. We were so deep into the creative process that it was almost painful to step back and downsize the project. In hindsight, I would’ve supported Steven as a co-producer rather than a client. Even after four years, I still believe in the potential of Polae to captivate a wide audience with its 360-degree interactive crime story.
Despite the bumps in the road, Steven and I have remained close, always keeping an eye on each other’s work. I love the guy and his ideas so much that I’d jump at the chance to work with him again.
So, let me tell you about Polae. It’s an interactive sci-fi series for mobile devices that follows Adelaide Engelsen and her company through a future history filled with cryptic articles, social media posts, and multimedia breaking news experiences. Viewers make decisions that ultimately determine their chances of survival when they come face-to-face with the artificial intelligence shaping the future.
Steven even created a noise ramp in StyleGAN for the film. It starts as a chaotic mess of color and gradually reveals the participant’s face. The real-time machine learning system and ML special effects were all developed by Steven himself, using generative adversarial networks hosted on AWS servers.
During the Polae experience, participants are asked to take a photo of themselves. The photo is then sent to a generative ML model, which creates five uncannily similar human faces. As the storyline unfolds and the world teeters on the brink of disaster, participants are faced with a dilemma: would they accept living as one of these recreated human forms in a post-Earth world if their original biological form couldn’t be perfectly replicated?
In the end, working with Steven and the incredible Mei-Ling Wong on this project has been a whirlwind of creativity, ambition, and friendship. Polae is the kind of project that will leave a lasting impression on anyone who experiences it, and I’m proud to have been a part of it.