About the Mind Enterprises Balcony AI Video
The Mind Enterprises balcony AI video generator recreates the viral balcony DJ clip with you and a friend in it. Two people stand behind the decks on a sunny balcony above a yellow city façade: one mixes with headphones on while the other dances beside the decks.
Upload two Character Sheets and the AI replaces both performers while keeping the original take. Your photos supply who appears in the video: face, hair, facial hair, body build, outfit and accessories. The reference clip supplies everything else: the performance, gestures, fixed camera framing, the balcony and buildings, the afternoon sunlight, the timing and the synchronised soundtrack.
How to make a Mind Enterprises balcony AI video
- Upload Image 1 — the DJ on the left. Add a Character Sheet of the person who should replace the DJ at the decks. One person per sheet: a front portrait, a front full-body shot and a back full-body shot give the model the most to work with.
- Upload Image 2 — the dancer on the right. Add a separate Character Sheet for the dancer beside the decks. Order matters: Image 1 is always the DJ on the left and Image 2 is always the dancer on the right, in every frame.
- Choose length and resolution. Pick 15 seconds for a quick loop or 30 seconds for the longer groove, then 480p for a cheaper preview or 720p for a sharper result. The credit cost is shown before anything is submitted.
- Generate, preview and download. Press Generate. Your video appears in your generations with a live progress status, and the finished vertical MP4 with the original audio is ready to watch and download there.
How the template works
The template uses motion transfer. Rather than pasting a face onto the original performer, the AI rebuilds each person from your Character Sheet and drives them with the exact motion of the source clip, so the result holds up when the performers turn, move out of frame or get close to the camera.
Each photo has one job and the mapping never changes: the first photo always replaces the DJ on the left, the second always replaces the dancer on the right. Nothing about the original performers is meant to carry over, including their hair, facial hair, sunglasses, clothes or jewellery.
What the AI keeps and what it changes
- Kept from the source video: the performance and gestures, the fixed camera framing, the balcony, buildings and sky, the sunlight, the DJ decks and headphones, the timing and the original audio.
- Taken from your Character Sheets: face, gender presentation, skin tone, body build, hairstyle and hair colour, facial hair, clothing, footwear and accessories.
- Never mixed: identities are locked to their side, so the DJ never borrows features from the dancer and the two never swap places mid-clip.
15 or 30 seconds
Both lengths are one continuous take in the original vertical format, starting after the headphones go on. Each length uses its own reference clip, so the 30-second option renders a genuinely longer performance rather than a stretched one.
Motion transfer is billed on the length of the reference, so 30 seconds costs twice as much as 15 seconds at the same resolution. The price table is below.
Tips for the best result
- Use one person per Character Sheet. Group shots make it harder to keep each identity on the right performer.
- Sharp, evenly lit photos that show the face and the outfit work best. Dress each sheet the way that person should look in the video.
- Keep the upload order: Image 1 is the DJ on the left, Image 2 is the dancer on the right.
- Start with 15 seconds at 480p to check your casting before rendering a longer or sharper version.
- Only upload photos of people who have agreed to appear in the video.
Length, resolution and credit cost
| Duration | 480p | 720p |
|---|---|---|
| 15s | 4,055 credits | 8,683 credits |
| 30s | 8,109 credits | 17,366 credits |
Credits are charged once when you start the render and refunded automatically if the generation fails or is blocked by moderation.
Frequently asked questions
How many photos do I need?
Exactly two. The first photo replaces the DJ on the left and the second replaces the dancer on the right.
Which performer does each photo replace?
Image 1 replaces the left performer, the DJ at the decks wearing headphones. Image 2 replaces the right performer, who dances beside the decks. The mapping stays the same for the whole video.
What stays the same as the original video?
The performance, gestures, camera framing, balcony setting, sunlight, timing and original audio. Only the two people change.
What is the difference between 15 and 30 seconds?
Each option uses its own reference clip from the same continuous take, so 30 seconds is a longer performance, not a slowed-down one. It uses twice the credits because motion transfer is billed on the length of the reference.
Does the video keep the original music?
Yes. Both reference clips carry the original synchronised audio, and the performance stays in time with it.
Why does a face, hairstyle or beard sometimes look off?
With fast movement, head turns or partly hidden faces, the model can occasionally drift toward the original performer's look for a few frames. Clear, well-lit Character Sheets of one person each reduce this. Try 15 seconds at 480p first to check a casting.
Is this an official Mind Enterprises tool?
No. This is an independent AI template. It is not affiliated with, endorsed by or sponsored by Mind Enterprises or anyone involved in the original video.
What happens if my generation fails?
The error is shown on your generation, and the credits for that render are returned to your balance automatically.
Discover more templates in the trends gallery or explore AI video effects.
The Mind Enterprises Balcony AI Video is an independent AI template. It has no official affiliation with, and is not endorsed or sponsored by, Mind Enterprises or the creators of the original video. The artist's name is used only to describe the source clip. Only upload images you have the right to use.