What happens in the Manhattan FPV drone video, second by second?
The whole video is one continuous FPV drone shot with no cut. Your photo is the end of the flight:
| Time | The camera | You | Pet |
|---|---|---|---|
| 0 to 2 s | High above the Manhattan skyline, the Art Deco tower with the spire in front, then it pitches down. | Not in view yet. | Not in view yet. |
| 2 to 5 s | A fast dive between the skyscrapers, then low over a long avenue and the yellow taxis. | Not in view yet. | Not in view yet. |
| 5 to 7 s | It crosses the curb and follows a stone facade along the sidewalk, slowing down. | You appear ahead on the sidewalk. | Your pet appears ahead on the sidewalk. |
| 7 to 10 s | A gentle hover in a close-up from the chest up. | You sip a coffee, lower the cup and look into the lens. | Your pet sniffs a coffee cup, looks up into the lens and tilts its head. |
How it works: this is an AI drone video from photo alone. The best AI video models available make the flight straight from your photo, with no in-between picture, so your face stays close to the photo. You get 10 seconds, vertical 9:16, 720p, with the wind and street sound the model makes for the flight. A run takes a few minutes, and the exact price shows on the Generate button before you spend anything. Every run is a new take, so the route and the buildings change a little each time.
