Hideaway Trend: Do the Viral Kiesza Dance From Your Photo

Strictly replicate the camera movements and actions of the character in the video. Replace the character in the video with the person from the uploaded image. Keep actions, poses, facial expressions and rhythm. Preserve the person's original facial features, maintain character consistency, no distortion, no glitches, no artifacts. The person dances in the place from the uploaded image, in the clothes from the uploaded image, not the multicolor striped top, the gold necklace or the black ruffled skirt of the dancer in the video, and keeps their own hair from the uploaded image, not long straight blonde hair. Follow the dancer's moves and timing exactly: leaning in very close to the camera and singing a long 'ooh... ah' with the eyes closed and big, dramatic faces, stepping back and tapping both hands on the chest to the beat, swinging the arms side to side, spinning around with the back to the camera and a hair flip, turning back with a wide-eyed stare, then singing to the camera again with both hands up by the face. Keep the body in view from the head to the hips, with natural balance. Exactly one person, no other people. One continuous shot, no cuts, no text, no captions, no logos.

Create a photorealistic vertical phone video of two friends doing a viral dance trend in an open lot at sunset: one dances, the other films and pops into the camera to sing. Never write the words of this prompt, photo labels or file names as text anywhere in the video. The dancer is the person in [Image 1], 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 and shoes from [Image 1]. The friend who films is the person in [Image 2], 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]. Never blend their faces. Each face stays the same from the first frame to the last, so their friends recognize them at once. If an image shows an animal, that performer is that animal, with its own face, fur and markings, not a person. The place: a wide, empty gravel lot at golden-hour sunset, warm orange light, long shadows, a pale clear sky. Behind the dancer: plain beige five-storey apartment blocks, two parked blue farm tractors, parked cars far back. ONE CONTINUOUS HANDHELD PHONE SHOT, filmed by the friend from [Image 2], NO CUTS. 0-3 s: the dancer, in the middle of the lot, seen head to knees, breaks into a loose, playful, wild dance: shoulder bounces, hip sways, arms flung wide, a goofy turn. 3-4.5 s: the friend turns the phone to their own face: the face from [Image 2] fills the lower half of the frame, very close to the lens, singing a long 'ooh... ah!' with big, dramatic faces, the dancer still dancing small behind. 4.5-8 s: back to the dancer, who dances harder: a spin, arms up, a stomp, a hair flip, a big grin. 8-9.5 s: the friend's face pops in close to the lens again, eyes wide: 'ooh... ah!' 9.5-12 s: the dancer ends with both arms thrown up and a laugh, the sunset glowing behind. VISUAL STYLE: photorealistic, like a real phone video, natural golden light, slightly shaky handheld, 9:16 vertical. Exactly two people, one head each, on their own shoulders. NO morphing faces, NO extra people, NO distorted hands. NO logos, brand names or text on the tractors, the buildings or anywhere, no subtitles, no watermark. Audio: an upbeat deep-house dance beat with a bright piano, the friend singing 'ooh... ah!' on top, light wind outdoors. No other words.

Image 1 replaces the cat who dances upright on its hind legs in the middle of the gravel lot, in every shot. Image 2 replaces the cat whose face pops up very close to the camera and sings, in every shot. Use image 1 and image 2 as the appearance references for the two pets: keep each pet's face shape, ears, fur colors, fur length, markings, body proportions and role consistent; do not swap roles or merge identities. SAME IDENTITY. FIXED ROLES. NO ROLE SWAP. Each pet keeps its own head shape, ears, fur length and coat from its own image, not those of the cats in the video. Each pet keeps the natural body volume and proportions of its own animal: never flattened, squashed, stretched or merged with anything. If an image shows a dog, it is that dog with its own face, ears and coat, up on its hind legs to dance or with its face close to the lens. If both images show the same pet, it plays both roles. Keep the gravel lot, the sunset light, the apartment blocks, the parked tractors, the dance moves, the motion, the timing and the camera exactly as in the video. No people. No text, no logos.