This is how it started: 3 photos shot by iPhone under natural lighting condition as data to feed for gen AI model.

I used Google Nano 🍌 to create a storyboard with key frames based on my vision. Then I picking music, adjust pacing, how the car should move and what could be added to make the scene more fun. I also thought about how the transitions should be. I had already imagined most of the film in my head, then I looked for ‘luck’ during the process.

I used 🍌 + Higgfields + VSCO + CapCut. These are such powerful and handy toolkits for making a film prototype. I haven’t used any upscaling techniques because I don’t think it’s necessary since this is still an experiment. So, I didn’t spend all my credit and $ on quality enhancement.

ZER0 / AI driven production