The balcony club dance AI video turns two portraits into a 15-second vertical scene: one person dances on the club floor while another watches from a balcony above. The role order matters. Photo 1 becomes the club dancer, and Photo 2 becomes the balcony watcher. You can create the complete scene with the balcony club dance AI video generator on imageat.
This tutorial covers the live two-photo workflow, the source images that work best, the current output choices, and the checks to make before posting. It is a different template from the retro balcony DJ performance: this version is built around a watcher-and-dancer story, a 9:16 frame, and a fixed 15-second sequence.
What the balcony club dance template creates
The template combines identity preparation, motion transfer, and a soundtrack replacement step into one workflow. Your uploads provide the two identities and outfits; the reference sequence controls the camera, choreography, club setting, and timing.
The current generator asks for:
- Photo 1 — club dancer: the person who appears on the dance floor.
- Photo 2 — balcony watcher: the person shown looking down from the balcony.
- Duration: 15 seconds.
- Format: vertical 9:16 video.
- Quality options: 480p SD or 720p HD.
- Audio: the trend soundtrack is added in the final workflow step.
The live page currently displays 201 credits for 480p and 426 credits for 720p. Product settings and credit costs can change, so treat those numbers as a snapshot and confirm the button total before generating.
Choose the right two photos
The template can prepare an identity sheet for each person, but it still needs clean visual evidence. A sharp source photo gives the model better information about the face, hair, clothing, and body proportions.
Use one clearly visible adult per photo. Choose images where the intended outfit can be seen, because the live upload instructions specifically ask for each person's outfit to be visible. A waist-up or full-body photo usually gives the dance role more useful information than a tightly cropped face.
For the best match between the two inputs:
- use photos with similar sharpness and lighting
- avoid strong beauty filters or motion blur
- keep eyes, hairline, and jawline unobstructed
- avoid hands covering the face
- choose a clear garment silhouette and recognizable colors
- use files you own or have permission to animate
If a good photo is too small, use the image upscaler before uploading. Keep the enhancement restrained. Oversharpening can create halos around hair and artificial skin texture that becomes more obvious in motion.

Assign Photo 1 and Photo 2 correctly
A common mistake is reversing the uploads. The interface labels the roles clearly, but it helps to rename the files before opening the generator:
01-club-dancer.jpg02-balcony-watcher.jpg
Photo 1 should show the person you want on the club floor. Photo 2 should show the person you want above the scene. Reversing them does not merely change a small detail; it swaps the characters' narrative positions.
Think about wardrobe as part of the role. A flexible, clearly defined outfit is useful for the dancer because the body moves through the frame. The watcher can work well with a clean waist-up portrait, but visible shoulders, arms, and clothing still help the model maintain continuity around the balcony rail.
Step-by-step: make the balcony club dance AI video
- Open the balcony club dance AI video generator.
- Watch the example so you understand the balcony-to-club structure and camera direction.
- Upload the club dancer as Photo 1.
- Upload the balcony watcher as Photo 2.
- Confirm that each preview shows the correct person and intended outfit.
- Select 480p for a lower-cost test or 720p for the sharper final option.
- Check the current credit total displayed on the Generate Video button.
- Start the generation and wait for all workflow steps to finish.
- Preview the entire 15-second result with sound.
- Download only after checking faces, hands, clothing, balcony rails, crowd details, and audio timing.
A lower-cost first pass is useful when you are unsure about role order or source-photo quality. Once the identities and outfits hold together, the higher-quality option is more appropriate for a final social upload.
How to improve identity consistency
Start with a readable face
Both people should remain recognizable when their faces become smaller or change angle. Use photos with even light across the face and enough resolution to show the eyes, nose, mouth, and hairline. Avoid extreme side profiles for the first attempt.
Show enough of the outfit
The workflow is designed to preserve each person's appearance as well as identity. A face-only crop leaves the model guessing about the torso and clothing. Show the intended top and, for the dancer, as much of the body as practical.
Avoid conflicting scene cues
A simple background is easier to interpret than a crowded photo with multiple people, mirrors, or strong shadows. The template creates its own club environment, so your source does not need to be taken in a nightclub.
Change the source before increasing resolution
Higher resolution can make a successful result look cleaner, but it cannot repair a hidden face, incorrect role order, or ambiguous body pose. When identity drifts, try a clearer source photo before paying for another higher-quality generation.
What to inspect before you publish
Faces and hair
Pause when the camera moves closer and when the scene changes between watcher and dancer. Check that the eyes, jawline, hair length, and hair color remain stable. Minor frame-to-frame variation is normal, but neither person should turn into a visibly different identity.
Hands and body movement
Dance motion can expose merged fingers, disappearing wrists, sliding feet, or sudden limb changes. Review the fastest movement at reduced playback speed. A single distracting frame may be noticeable because short videos are often replayed automatically.
Balcony and club continuity
Use architectural lines as checkpoints. The balcony rail, walls, lights, and crowd should remain spatially coherent as the camera changes perspective. Watch for railings that melt into hands or background faces that pull attention away from the main subjects.
Clothing consistency
Look for changing sleeves, shifting necklines, unstable jewelry, or colors that change between shots. If clothing is the main problem, select source photos in which garment edges and texture are clearly visible.
Audio and ending
Listen through the complete export. Confirm that the soundtrack is present, major movement feels aligned with the edit, and the clip does not begin or end abruptly. Platform previews can mute by default, so test the uploaded draft with sound enabled before making it public.
Posting the result to Reels, TikTok, and Shorts
The finished 9:16 video is already shaped for vertical feeds. Avoid unnecessary crops or repeated exports. Upload the clean downloaded file directly whenever possible.
A concise caption should explain the two-role setup rather than pretend the scene was a real event. For example: “Two photos, two roles: club dancer below and balcony watcher above.” Add a clear disclosure when the realistic scene could otherwise confuse viewers.
Useful publishing checks include:
- choose a cover frame with a clear face and readable action
- keep interface text away from platform button areas
- do not cover the dancer's face with a large title
- preview the post on a phone before publishing
- confirm that audio survived the upload
- use only identities and audio you have permission to publish
For other movement-first concepts, explore the AI dance video generator. If you want to replace the fixed choreography with your own camera and action instructions, use the more flexible AI video generator. You can also browse current imageat trends for other one-photo and multi-photo templates.
Balcony club dance vs. the retro balcony dance template
The names sound similar, but the scene design is different. The balcony club dance workflow places a watcher above a club dancer and always produces a 15-second vertical scene from two role-specific photos. The retro balcony dance template is structured like a coordinated music performance and offers its own reference sequence and settings.
Choose the template by story, not by the shared word “balcony.” Use this page when you want the clear watcher-versus-dancer composition. Use the retro performance workflow when you want two people participating in the same balcony routine.
Responsible use
Only animate people who have agreed to the use of their photos. Do not use the effect to fabricate evidence, impersonate someone, suggest a relationship that does not exist, or imply that a real person attended or endorsed an event.
The example is AI-generated, and your post should not be framed as documentary footage. If the soundtrack has restrictions outside the template or platform, confirm that you have the necessary rights for commercial, advertising, or client work.
Frequently asked questions
How many photos do I need?
You need two photos. Photo 1 is the club dancer, and Photo 2 is the balcony watcher.
How long is the generated video?
The current template produces a 15-second video in a vertical 9:16 format.
Does the result include sound?
Yes. The live workflow includes a final soundtrack replacement step. Always play the downloaded file and the platform draft with sound to verify that audio is present.
Which quality option should I choose?
Use 480p for a lower-cost identity and role-order test. Choose 720p when the test works and you are preparing the final version. Check the current credit total on the live page because pricing can change.
Can I use group photos?
A group photo can make identity selection ambiguous. Use one clearly visible person per upload for the most predictable role assignment.
Can I change the choreography?
The trend page uses a fixed reference sequence. For custom action, location, camera movement, or timing, start with the AI video generator or a dedicated motion transfer workflow.
Why does one face change during the video?
Identity drift is more likely with blur, heavy filters, extreme angles, hidden facial features, or low-resolution inputs. Replace the weaker source photo and test again before increasing quality.
Create your balcony club dance video
Prepare two clear portraits, place the dancer in Photo 1 and the watcher in Photo 2, and review the full result with sound before posting. The strongest output comes from clear identities, visible outfits, correct role order, and a careful final quality check.
When both photos are ready, open the balcony club dance AI video generator on imageat, confirm the live quality and credit options, and generate your 15-second vertical scene.
