APT AI Video Generator: Turn Two Images into a Duo Video
Put yourself and a friend in a playful pink-studio performance. Explore our APT AI video generator workflow and copy-ready prompts to turn two character images into an APT-inspired duo video. Assign one person to the singer and the other to the drummer, then use a reference clip to guide gestures, interaction, and camera cuts. Create your own APT AI dance video with recognizable faces and a clear role for each performer.
Use images of yourself or authorized adults. Choose a generation mode that supports two character images and a video reference; available inputs and output settings depend on the selected model.
What Is the APT AI Video Trend?
The APT AI video trend reimagines the ROSÉ and Bruno Mars performance with a new duo in a recognizable pink studio. Two images define the cast, while a selected video reference guides the singing, drumming, playful gestures, and shot sequence. Unlike a generic APT dance video, a reference-based version aims to keep each replacement attached to one performer across the scene.
Two Images, Two Consistent Characters
Use Image 1 for the singer and Image 2 for the drummer in the prompts below. Preserve each person's face, hairstyle, skin tone, outfit, and body proportions. Bind identities to these roles throughout the sequence, even when the performers change sides or appear alone.
Pink Studio, Reference-Guided Performance
Let the reference clip guide gesture timing, drum actions, pink-studio scenery, and camera cuts. Keep only the props and interactions visible in your chosen segment. Review short test renders for hand distortion, face drift, and role confusion before generating a longer version.
How to Make an APT AI Video from Two Images
Step 1: Upload Two Character Images
Open Reference-to-Video AI and choose a mode that accepts two character images and a video reference. Upload one clear, well-lit image per person, ideally showing the face, torso, and arms. In the prompts below, Image 1 takes the singer role and Image 2 takes the drummer role.
Step 2: Add a Reference and Prompt
Copy an APT AI video prompt below and add the performance segment you want to follow. Check the singer and drummer mapping against that clip. Select a supported duration matching the segment and preserve its framing; use 9:16 only when both performers and key props fit the crop.
Step 3: Generate, Review, and Download
Generate your APT-inspired AI video and check both characters through every cut, including solo shots. Look for identity swaps, drifting faces, distorted hands, or changing outfits. Refine the role instructions if needed, then download a usable take and finish the soundtrack separately when required.
Copy-Ready APT AI Video Prompts for Your Duo
Two Images + One Performance Reference
Classic APT Duo AI Video Prompt: Use @Image1 as the singer and @Image2 as the drummer in @Video1. Follow these performer roles throughout, never their changing left/right positions. SAME IDENTITY: preserve each uploaded person's facial features, skin tone, hairstyle, body proportions, and outfit in every shared and solo shot. Recreate only the selected reference segment, following its singing gestures, drum movements, expressions, interactions, action timing, camera angles, movement, and cut order. Keep the pink studio, lighting, instruments, and props shown in @Video1. Replace the singer in every singer close-up and the drummer in every drummer shot. Natural hands, stable instrument contact, grounded feet, and consistent clothing. Exactly two assigned performers overall; retain the reference's person count in each shot. NO DUPLICATE, no identity swaps, face blending, original-performer reappearance, extra people, extra limbs, or clothing flicker. NO NEW SCENES. Match the reference framing and selected duration; handle music separately.
Two Adult Images + One Performance Reference
Couple APT AI Video Prompt: Recast the APT-inspired performance in @Video1 with two authorized adults: @Image1 follows the singer and @Image2 follows the drummer. SAME IDENTITY across every cut: retain their own faces, hair, skin tones, clothing, and gender presentation. Copy the reference's playful expressions, shared gestures, drumming, and any affectionate interaction exactly where it appears; do not add kisses, embraces, or new actions. Preserve the pink stage, props, lighting, shot sequence, camera motion, and timing from the selected segment. In a solo shot, show only the person assigned to that performer. Keep the same role when either person changes screen position. Maintain separate faces, natural hands, stable feet, and consistent outfits. NO DUPLICATE, no blended identities, added people, original faces, warped fingers, sliding feet, or sudden background changes. NO NEW SCENES. Use output settings matching the reference crop and segment length; add the soundtrack separately if needed.
Two Friend Images + One Performance Reference
Best Friends APT Dance Video Prompt: Put the friend in @Image1 into the singer role and the friend in @Image2 into the drummer role of @Video1. SAME IDENTITY: preserve each friend's recognizable face, hairstyle, outfit, skin tone, and body proportions through wide shots, close-ups, and cuts. Follow the reference's energetic duo performance, including only the gestures, drum actions, glances, poses, and interactions present in the selected segment. Keep action timing, camera framing, camera movement, pink-studio background, instruments, and lighting aligned with @Video1. Do not assign identities by left/right position. Replace both original performers in all relevant shots, including individual shots. Exactly two distinct characters overall; no extra audience. Natural joint motion, separate hands, stable props, grounded steps, and no outfit flicker. NO DUPLICATE, no role swapping, face merging, extra limbs, or new choreography. NO NEW SCENES. Match the chosen reference duration and aspect ratio; finish music separately.
Two Character Images + One Performance Reference
Original Character APT AI Video Prompt: Use two original humanoid characters from @Image1 and @Image2 to reinterpret the performance in @Video1. @Image1 always follows the singer; @Image2 always follows the drummer, regardless of screen position. SAME IDENTITY: retain each character's distinct face, silhouette, hairstyle, colors, and costume throughout every shared and solo shot. Transfer the reference's gesture sequence, expressions, drumming, interaction timing, camera motion, framing, and cuts without inventing new actions. Keep the pink studio and the instruments and props visible in the selected segment. Preserve the characters' reference design while fitting them naturally into the stage lighting. Maintain anatomically coherent hands, stable drumstick contact, grounded feet, and consistent costumes. NO DUPLICATE, no character fusion, identity swaps, extra performers, original human faces, extra limbs, or background flicker. NO NEW SCENES. Match the reference segment length and framing; plan audio separately.
Explore the Hottest AI Features and Effects
FAQs About the APT AI Video Generator
1. What is an APT AI video generator?
An APT AI video generator uses character images to create a duo performance inspired by the ROSÉ and Bruno Mars music video. In a reference-based workflow, two images supply the identities while a video clip guides movement, scenery, and camera cuts. Results vary with the model, inputs, and selected reference.
2. How do I make an APT AI video from two images?
Choose a reference-to-video AI generator mode supporting two character images. Upload one person per image, assign the singer and drummer roles, and add a short performance reference. Copy a prompt from this page, choose supported output settings, then review faces, hands, role continuity, and audio before downloading.
3. Can I create an APT dance video without a reference clip?
You can describe a pink studio and playful duo choreography with text, but that will not reliably reproduce a specific sequence of actions or camera cuts. For a closer APT-inspired performance, use a compatible video-reference workflow. With only a first-frame image, aim for a simpler scene and shorter motion sequence.
4. Does an APT AI video include the original song and lip-sync?
Not necessarily. An APT AI video workflow may follow visible mouth movements without embedding the original song or producing accurate lip-sync. Check the selected model's audio options and preview the result. If needed, add suitable music in your editor or publishing platform and adjust its timing to the performance.
5. How do I prevent character swaps after camera cuts?
Use clear individual images and specify SAME IDENTITY for each character. Bind Image 1 to the singer and Image 2 to the drummer throughout, including solo shots; avoid changing left/right assignments. Add NO DUPLICATE, then review cuts and occlusions. Shorten a difficult segment or improve the references if identities drift.
Explore More AI Dance Videos, Duo Trends, and Character Prompts
Upload. Assign Roles. Generate.