Coffee Making AI Video: Put Anyone, Any Pet or Any Character in the Kitchen

Replace the tall knight in black armor behind the counter with the whole subject in [Image 1], and replace the small armored kitten on the counter with the whole subject in [Image 2]. Each photo may show a person, pet, toy or character, and that whole body takes the place: a person keeps their face and clothes, a pet keeps its fur and paws, a toy keeps its shape. Nothing of the knight or the kitten remains. Keep the wooden kitchen, the red mug, the spilled grounds, the milk splash and the fixed camera. [Image 1] spoons the coffee and holds the mug. [Image 2] watches from the counter. Edit [Video 1] in place. Use each uploaded image ONLY as the identity reference for its assigned role. A photo is not a room, pose, camera or lighting reference. Preserve the complete original performance, including frame order, duration, camera, lighting, location, body positions, timing and any on-screen text already in [Video 1]. Do not restage the clip or create new shots. Do not add new text, logos or a watermark. Produce a silent video; the original reference soundtrack is added separately. The reference video is about 15 seconds. Preserve its full action and timing without retiming, extending, adding or removing shots.

Replace the adult on the right, who spoons coffee and spills the grounds, with the adult in [Image 1], and the adult on the left, who pours the milk, with the adult in [Image 2]. Both keep their own faces, hair and clothes and stay calm while the grounds and milk go everywhere. Keep that kitchen, the camera and the timing. Exactly two people. Never swap the two faces. Edit [Video 1] in place. Use each uploaded image ONLY as the identity reference for its assigned role. A photo is not a room, pose, camera or lighting reference. Preserve the complete original performance, including frame order, duration, camera, lighting, location, body positions, timing and any on-screen text already in [Video 1]. Do not restage the clip or create new shots. Do not add new text, logos or a watermark. Produce a silent video; the original reference soundtrack is added separately. The reference video is about 14 seconds. Preserve its full action and timing without retiming, extending, adding or removing shots.

Replace only the tall knight in black armor with the whole subject in [Image 1]: a person, pet, toy or character, with its own body, face or fur, and its own clothes or surface. The small armored kitten on the counter stays exactly as in [Video 1]. Nothing of the knight remains. Keep the wooden kitchen, the red mug, the spilled grounds, the milk splash and the fixed camera. [Image 1] does the knight's coffee-making moves in the way its own body allows. Edit [Video 1] in place. Use each uploaded image ONLY as the identity reference for its assigned role. A photo is not a room, pose, camera or lighting reference. Preserve the complete original performance, including frame order, duration, camera, lighting, location, body positions, timing and any on-screen text already in [Video 1]. Do not restage the clip or create new shots. Do not add new text, logos or a watermark. Produce a silent video; the original reference soundtrack is added separately. The reference video is about 15 seconds. Preserve its full action and timing without retiming, extending, adding or removing shots.

Replace only the dog on the left, who pours from the black kettle, with the adult in [Image 1]. [Image 1] is a human head and body in their own clothes, leaning on the steel counter at adult size, not a dog and not shrunk to the dog's size. The dog on the right stays exactly as in [Video 1], small, watching and stirring. Keep the sage-green kitchen, the white mug, the spilled grounds, the knocked-over milk and the innocent stare at the end. Exactly one person and one dog. Edit [Video 1] in place. Use each uploaded image ONLY as the identity reference for its assigned role. A photo is not a room, pose, camera or lighting reference. Preserve the complete original performance, including frame order, duration, camera, lighting, location, body positions, timing and any on-screen text already in [Video 1]. Do not restage the clip or create new shots. Do not add new text, logos or a watermark. Produce a silent video; the original reference soundtrack is added separately. The reference video is about 15 seconds. Preserve its full action and timing without retiming, extending, adding or removing shots.

Replace the dog on the left with the adult in [Image 1] and the dog on the right with the adult in [Image 2]. Both are grown adults in their own clothes, leaning on the steel counter at adult size. [Image 1] pours from the black kettle. [Image 2] stirs the white mug. Keep the sage-green kitchen, the milk spill and the timing. No dogs remain. Exactly two people. Never blend the two faces. Edit [Video 1] in place. Use each uploaded image ONLY as the identity reference for its assigned role. A photo is not a room, pose, camera or lighting reference. Preserve the complete original performance, including frame order, duration, camera, lighting, location, body positions, timing and any on-screen text already in [Video 1]. Do not restage the clip or create new shots. Do not add new text, logos or a watermark. Produce a silent video; the original reference soundtrack is added separately. The reference video is about 15 seconds. Preserve its full action and timing without retiming, extending, adding or removing shots.

Replace only the dog on the left with the pet in [Image 1]. The pet keeps its species, face, ears, fur and paws and pours from the black kettle. The dog on the right stays exactly as in [Video 1]. Keep the sage-green kitchen, the white mug, the milk spill and the paw-over-the-face ending. Exactly two animals. Animal paws, never human hands. Edit [Video 1] in place. Use each uploaded image ONLY as the identity reference for its assigned role. A photo is not a room, pose, camera or lighting reference. Preserve the complete original performance, including frame order, duration, camera, lighting, location, body positions, timing and any on-screen text already in [Video 1]. Do not restage the clip or create new shots. Do not add new text, logos or a watermark. Produce a silent video; the original reference soundtrack is added separately. The reference video is about 15 seconds. Preserve its full action and timing without retiming, extending, adding or removing shots.

Replace the dog on the left with the pet in [Image 1] and the dog on the right with the pet in [Image 2]. Each pet keeps its own species, face, ears, fur and paws. [Image 1] pours from the black kettle. [Image 2] stirs the mug. Keep the sage-green kitchen, the milk spill and the timing. Exactly two animals. Never blend the two pets or give them human hands. Edit [Video 1] in place. Use each uploaded image ONLY as the identity reference for its assigned role. A photo is not a room, pose, camera or lighting reference. Preserve the complete original performance, including frame order, duration, camera, lighting, location, body positions, timing and any on-screen text already in [Video 1]. Do not restage the clip or create new shots. Do not add new text, logos or a watermark. Produce a silent video; the original reference soundtrack is added separately. The reference video is about 15 seconds. Preserve its full action and timing without retiming, extending, adding or removing shots.