Smooth Criminal Dance AI: The Viral Routine From One Photo

Track the exact camera movement, framing, facial expressions, and body movements from the video. Keep the face, hairstyle, body, clothing, and overall appearance exactly the same as in the reference image, with no changes or distortion. The performer is the person from the reference image instead of the man in the video, standing upright in the place from the reference image. Keep every move of the dance routine in time with the beat: the hat-brim tilt, the side glide, the kick and the spin, from the first count to the final pose. Output in 9:16 vertical format. The person keeps natural human anatomy: two arms and two hands with five fingers each, two legs, no extra limbs, even in the fast spins and kicks. No cuts, no text, no captions or labels on screen. No light gray cap, no white T-shirt and no baggy light blue jeans unless the reference image shows them. Exactly one person, no extra people.

Recreate the reference video [Video 1] shot for shot: the same paved path in front of a row of green trees under a blue sky, the same low fixed phone camera that films both dancers' whole bodies from the front, the same framing, timing and dance moves, one continuous take with no cuts. The ONLY change: the dancer on the LEFT is now the person in [Image 1], and the dancer on the RIGHT is now the person in [Image 2]. Use [Image 1] as adult Person A, on the left, and [Image 2] as adult Person B, on the right. Keep each person's face, hairstyle, outfit and body proportions consistent; do not swap sides or merge identities. Person A: [Image 1] replaces the woman on the left with long brown hair, in EVERY frame: copy the exact face from [Image 1] (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). They wear the clothes from [Image 1]. NO navy tank top, no wide white trousers and no long brown hair unless [Image 1] shows them. Person B: [Image 2] replaces the woman on the right with long blonde hair, in EVERY frame: copy the exact face from [Image 2] (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). They wear the clothes from [Image 2]. NO navy top, no wide white trousers and no long blonde hair unless [Image 2] shows them. Both change: no one from the video stays, also in the spins and when only the back of a head shows. The same real people from the first second to the last, so their friends recognize them at once. Never morph, never swap the two faces, never drift back to the original dancers' faces. If [Image 1] or [Image 2] shows only the head or the upper body, invent the rest so it matches naturally. The two dance the same routine in sync, move for move as in the reference: the hat-brim tilt (a quick hand flick at the forehead), sharp arm swings, the side glide, hip and knee pops, the kick to the side, quick spins, and the pose at the end. One head on each person, five fingers on each hand, no extra limbs. Exactly two people. No text, no captions, no logos, no watermark.

Recreate the reference video [Video 1] shot for shot: the same cozy living room with a beige carpet, wooden doors, framed pictures, potted plants and warm lamp light, the same low camera at the pets' height, the same cuts, light changes, framing, timing and dance moves. The ONLY change: the small dog in the black fedora who dances upright in front, on the right, is now the pet in [Image 1]. The small cat in the white fedora behind it, on the left, stays exactly as in the video. The lead dancer = the pet in [Image 1], whole body, in EVERY frame: the same species and breed, head shape, muzzle, eyes, ears, fur colors, fur length, markings and body shape as in [Image 1], with the natural body volume and proportions of that animal: never flattened, squashed, stretched or merged with anything, so its owner instantly recognizes it, also in the close-ups. It stands upright on its hind legs like a person the whole time. Never drift back to the dog of the original video: not the coat, not the big upright ears and not the face of the dog in the video, unless [Image 1] shows them. If [Image 1] shows a cat or another animal, that animal dances in the same place with the same moves. The dance, matching the reference shot by shot: 0-1.6 s: both stand upright in their fedoras, front paws at the chest, small hip sways. 1.6-3.5 s: the light dims; the lead pet raises one front paw to the brim of its black fedora and tilts it. 3.5-5 s: the light comes back; the lead pet holds the brim, then throws the fedora up and away; both hats come off. 5-15 s: without hats, both dance side by side: paw swings, side steps, hip sways and turns; the lead pet leans in close to the camera and back. The pet keeps real animal anatomy: furry forelegs ending in round paws covered in its own fur, never human hands, no bare skin, no fingers. One head only, two hind legs and two front legs, no extra legs or paws. Exactly two animals: the pet from [Image 1] and the cat of the video. No people, no text, no logos, no watermark.