Recreate the reference video [Video 1] shot for shot: the same music classroom with the cream walls, the strip lights and the row of guitars on the wall, the same black keyboard in front, the same light, camera angles, cuts, timing and moves. The ONLY change: [Image 1] replaces the cat who stands upright on the left and sings to the ceiling, and [Image 2] replaces the cat who stands at the keyboard on the right and plays. Each photo can show a person, a pet, a toy or a character: it takes that cat's place with its WHOLE body. [Video 1] gives only the place, the camera, the cuts and the moves: no cat from the video stays in any shot.
The singer on the left = [Image 1], in EVERY shot, from the very first frame. If [Image 1] shows a person: copy the exact face (the face shape, eye color and eye shape, eyebrows, nose, mouth, skin tone, age, hair color, hair length and hairstyle, and any beard, moustache, glasses or freckles) and the clothes from [Image 1]; a full-grown adult at their real height and build, the keyboard at their hips, never a child's size or proportions. If it shows an animal: the same species and breed, face, muzzle, eye color, ears, fur colors, fur length and markings as in [Image 1], with the natural body volume of that animal, upright on its hind legs, with real furry paws; its coat is not the coat of a cat in the video. If it shows anything else (a toy, a cartoon or 3D character, a figure or an object): that exact subject with its own shape, proportions, colors, material and art style; its own whole body replaces ALL of the cat, with no fur, cat ears or whiskers unless [Image 1] shows them; a plush stays soft, a blocky figure stays blocky. It sings to the ceiling, the mouth open on the long notes when it has one.
The pianist on the right = [Image 2], in EVERY shot, copied from [Image 2] in the same way. It plays the keys with its own hands, paws or limbs and wears the pianist's black wristwatch when it has an arm or a leg.
If anything still looks like a cat from the video and not like its photo, it is wrong. The same two in every frame, so their friends recognize them. Never morph or blend them: [Image 1] is only the singer, [Image 2] only the pianist.
The food bowl on the keyboard in front is a plain cream bowl with no picture, no print and no letters.
Shots, matching the reference:
Shot 1 [0-3.0 s]: close shot: the singer from [Image 1] on the left sings to the ceiling, head tipped back; the pianist from [Image 2] on the right turns to the singer and keeps the beat in the air, like a conductor.
Shot 2 [3.0-5.7 s]: hard cut, a wider shot along the keyboard: the pianist from [Image 2] plays; the singer from [Image 1] sings to the ceiling.
Shot 3 [5.7-14 s]: hard cut, a wide shot under the strip light: the pianist from [Image 2] plays and looks up at the singer; the singer from [Image 1] belts the long notes, head tipped back, eyes half closed.
Anatomy rules: one head each, on its own body. Exactly two performers, no extra people or animals. No dialogue throughout the entire video. No background music. No caption, no text, no logos, no watermark.
3124 characters