Put the cats from [Image 1] and [Image 2] in the places of the kittens in [Video 1], [Image 1] in the middle and [Image 2] on the right (one cat takes the middle spot), and keep every move of the cat dance in time with the beat, the sway, the wiggle and the paws up, from the first move to the last. Real cat anatomy: paws, never human hands. Output in 9:16 vertical format.
Recreate the reference video [Video 1] shot for shot: the same wet sand beach at night, the calm dark sea with the lit city skyline far behind, the tall grass on the left edge, the same low fixed phone camera close to the sand, one continuous take with no cuts, the same framing, timing and moves, the three dancers side by side in the same three spots. The ONLY change: two kittens are now the pets in the photos: middle (the studded black belt) = [Image 1], right (stands still) = [Image 2]. Both change: no kitten from the video stays in those two spots. The kitten on the left stays exactly as in [Video 1]. [Image 1] and [Image 2] are the sole identity references, one per dancer; never blend their looks. [Video 1] is for the moves, the camera and the beach only.
The middle dancer = the pet in [Image 1] in EVERY frame: its species, breed, face, eyes, ears, muzzle, fur colors, fur length and markings, never the face or the coat of the kitten of the video in that spot. It keeps the studded black belt, sized to its own body.
The dancer on the right = the pet in [Image 2] in EVERY frame: its species, breed, face, eyes, ears, muzzle, fur colors, fur length and markings, never the face or the coat of the kitten of the video in that spot. Nothing on it.
Each pet keeps the natural body volume and proportions of its own animal: never flattened, squashed, stretched or merged with anything. Each stands upright on its hind legs like the kitten it replaces, at the same size, and stays the same from the first second to the last, also when it turns its back to the camera, so its owner instantly recognizes it. If a photo shows a dog or another animal, that dancer is that animal, with its own ears, muzzle and tail.
The dance, matching [Video 1] second by second, on the beat. 0-1.5 s: the middle dancer faces the camera, front paws up high. 1.5-4.5 s: it turns its back to the camera, wiggles its hips and tail, turns back. 4.5-6 s: paws out to the sides, a sway. 6-9 s: paws up, then paws at the chest, swaying side to side. 9-12 s: it turns its back again, wiggles, turns around. 12-15 s: paws out, then paws up high. The dancer on the left keeps its back to the camera the whole time, paws out, swaying. The dancer on the right stands almost still and stares at the camera, exactly as in [Video 1].
Real animal anatomy: round furry paws, never human hands or fingers. One head, four legs and one tail each. Exactly three animals, no people.
No dialogue throughout the entire video. No background music. No text, no captions, no logos, no watermark.
2940 characters