We can all get a little obsessive at times. London Alley Director Courtney Phillips took the feeling to its cinematic ceiling in the video for REGARD’s dance hit “You” featuring Troye Sivan and Tate McRae. Through experimental use of deepfake technology, Courtney sets viewers on the edge of time and sanity to witness what can happen when your information is not thoroughly password protected.
“From the start, the question was how can we tackle a shoot where there’s not one, but three artists in different countries?” Courtney recalled, “this was my first time shooting remotely and it forces you to approach things from a different perspective.” Bringing on a specific set of constraints, the project required an aptitude for creative problem solving. “It was the tone of the track and the repetition of “you” in the lyrics that made me think of an obsession, a person totally losing themselves in the thought of someone else,” Courtney described. “I wanted to modernize the stalker concept. How in the age of social media and technology would one become someone else?” That’s when using deepfake clicked. The cutting edge and slightly off-putting technology replaces one’s likeness with another’s through deep learning, making it the perfect complement to the song, as well as the dispersed filming conditions.
Utilizing the highly sensitive technology in this way took a special expertise. Luckily, London Alley Post Production Supervisor Iván Ovalle had previous experience with the medium and was able to lend his knowledge to the project. “I was able to help guide Courtney as to how to capture footage that would work most effectively in the deepfake process,” Iván said. “There are limitations in terms of actor movement, lighting, and camera angles that had to be taken into consideration to get the best result.” Much of the creative storytelling was built in post production, folding each artist’s isolated footage into one manic narrative in a way that maintained perfect synergy.
The video reached across many minds within the London Alley ecosystem, relying on widespread collaboration, innovative thinking, and trust. “The implications of using bleeding edge machine learning technology is that the outcome will always be unpredictable,” noted London Alley Editor Luis Caraza, “Once our deepfake artist began to work, we wouldn’t be able to see the final shots until the very end of the process. We prepared backup plans in case we were dissatisfied with the results, but in the end they turned out just how we envisioned.”
The eccentric nature of the project was encouraged through its unique challenges, Courtney’s distinct style, and the expansive expertise at London Alley. Check out the mind-bending video below.