Metronomia

• 

  Metronomia

Inspiration

The subway has always fascinated me — a liminal space where strangers share fleeting moments, each immersed in their own world. I often wonder: what are they feeling, thinking, escaping? One day, I saw someone close their eyes as music filled their headphones, and for a moment, they seemed to transcend the noise. That was the seed for Metronomia — a poetic journey through music, emotion, and urban motion.

What it does

Metronomia is a short, AI-generated cinematic film that visualizes the inner worlds of subway passengers, as if the train ride were a vessel into their emotions. Each scene is built around the transformative power of music — from moments of connection to states of trance or introspective stillness — captured with dynamic camera movements and ultra-realistic rendering. It’s an ode to ordinary beauty, invisible ties, and the rhythm of city life.

How we built it

ChatGPT-4o – to generate, refine, and iterate highly cinematic prompts, narratives, and shot ideas. Minimax Hailuo 02 – for dreamlike, cinematic renderings with expressive facial realism. DaVinci Resolve – editing, grading, and sequencing final footage with a cohesive tone.

Challenges we ran into

Balancing realism and dreamlike elements without losing narrative clarity. Getting consistent quality across AI video models. Building emotional tension in 6-second sequences with no dialogue or voiceover. Maintaining visual continuity across shots while embracing stylistic freedom.

Accomplishments that we're proud of

Achieved a poetic, emotionally immersive visual experience using only AI-generated footage. Created cinematic moments that would be impossible to shoot in real life (e.g. passing through glass, orbital drones inside a train, floating passengers). Kept the tone grounded, sincere, and human despite using high-concept, tech-driven tools. Captured the universal feeling of being alone together — a quiet, shared urban intimacy.

What we learned

Precision in prompting is everything — film language can be translated to prompt language. AI tools are collaborators, not shortcuts. They require direction, storytelling, and iteration. Emotional resonance doesn’t require actors or dialogue — framing, motion, and rhythm can speak volumes. Music is a narrative vector. It shapes perception, timing, and emotional impact even in silent visuals.

What's next for Metronomia

Push further into surreal territory, using AI to bend urban reality without losing authenticity. Develop a gallery installation version for immersive projection in physical spaces.

Built With

• english