What happens in the CATEL dance video, second by second?
Both versions copy the same clip, one continuous take from a phone on the ground, so the camera looks up a little. Your photo takes the dancer's place, and the courtyard stays:
| Time | Human | Pet |
|---|---|---|
| 0 to 1 s | You stand in the sun, a hand at your hip, and step toward the camera. | Your pet stands up on its hind legs and steps in. |
| 1 to 2.5 s | You lean in to the low lens and point at the camera with a big smile. | It leans in and points with a front paw. |
| 2.5 to 5 s | A hip sway, then a low squat with your hips to the side and one arm out. | A hip wiggle and a low crouch, one paw out. |
| 5 to 6.5 s | Both hands go up and through your hair. | Both paws go up to its head and ears. |
| 6.5 to 9 s | A turn to the side with a point, a hip pop and small fists at the waist. | A turn, a paw point and a wiggle. |
| 9 to 11 s | A low squat, both hands flick down in time, and you sing along. | A low crouch and paw flicks, mouth open as if it sings. |
How it works: the best AI video models available edit the clip itself: they swap the dancer for the person or pet in your photo and keep the courtyard, the light and the camera. You get 11 seconds, vertical, at 720p, and the CATEL song goes back on after the run, so it plays over your video. The mouth follows the clip, so it is close to the words but not exact lip sync. A run takes 5 to 15 minutes, and the exact price shows on the Generate button before you spend anything. Every run is a new take, so results differ.
