Skip to main content
Dame Hombro AI

Video template

Dame Hombro AI Dance Template

Make yourself the lead in a Dame Hombro-inspired dance outside a red-lit venue. This template starts with one adult portrait and requests a new scene of shoulder bounces, claps and gestures toward a small crowd. The current output is a 15-second video at 720p, with generated instrumental and ambient audio. Choose vertical 9:16 or landscape 16:9 in the generator. The example introduces the mood; your photo supplies the person who takes center stage.

15 seconds720p9:16 / 16:91 photos
Scene reference video. Sample duration and audio may differ from a new generated result.

The scene

Vivid red light frames the entrance and stairs behind the performer. A portable speaker and a phone sit low in the composition while adult onlookers gather nearby. The requested waist-up view uses gentle handheld movement. The lead briefly turns toward the circle before facing the camera again, connecting a personal dance moment with the surrounding audience.

Prepare your photos

Choose one clear portrait of an adult who has agreed to appear. Keep the face, hairstyle and shoulders visible, with enough clothing detail to serve as a reference. There is no need to recreate the red lighting in your photo: the setting comes from the template. The crowd is generated separately, so a group photograph is not needed.

Use this template

One credit per video. Packs from $9.99. View pricing

What makes this dance scene distinctive

The main preset centers on a small exchange between the performer and onlookers. Energetic shoulder bounces establish the rhythm; claps and open palms make the response readable. The brief turn toward the crowd is part of the request, followed by attention returning to the camera. These actions describe the intended performance without assigning a gesture to every second.

The entrance and stairs provide a recognizable nighttime setting, while the speaker and phone remain background context near the bottom of the frame. This is a waist-up composition, so the template emphasizes shoulders, facial expression and hands rather than a detailed footwork routine. Gentle handheld framing is requested; a specific camera path is not prescribed.

The prompt asks for one continuous scene, with no cuts, added text, logos or watermarks. It also asks to keep the reference person recognizable and separate from the generated crowd. Those are generation instructions, not guarantees about every frame. Watch the complete clip to judge the transitions, the crowd interaction and whether the lead remains consistent.

Prepare a portrait for the red-lit scene

Keep useful detail. An original image with a clearly visible face gives more reference information than a small screenshot or a distant figure. Include the shoulders when possible. Avoid cropping through the chin or covering the eyes with a phone, sunglasses or hair.

Use readable lighting. Even daylight can be a useful starting point, despite the generated scene taking place at night. Strong red filters, harsh shadows and heavy smoothing may hide the features you want retained. Choose a photo where you can recognize the person easily before considering whether its colors match the sample.

Choose one person. This preset has one lead slot. Additional faces in a source image are not assigned separate crowd roles. If the point of your video is to recognize several friends, switch to the friends template and use its separate portrait slots. Confirm permission for every person whose photo you upload.

Three checks before another attempt

The face changes while turning. Compare the affected frames with the source portrait. A sharp, unobstructed face with even lighting may offer a clearer reference, but identity drift can still occur. Reconsider the photo before generating again; a different reference does not guarantee a correction.

Claps or hands look awkward. Review the movement as a sequence, especially where palms meet or separate. Hand anatomy and contact can vary in generated video. A photo showing more shoulder and upper-body context may help interpretation, without proving that the original crop caused the problem.

The framing differs from the sample. Confirm the selected preset and aspect ratio. Landscape and portrait compositions leave different room around the lead and crowd. The example is not transferred frame by frame. Review completed attempts in My videos before deciding whether another photo or a different scene better fits your idea.

Sunlit plaza solo: a daytime alternative

The sunlit plaza preset uses one adult lead photo. Warm orange and cream buildings, balconies and golden afternoon light replace the night entrance. The performer shifts weight, rolls each shoulder and makes relaxed hand gestures while approaching the camera. The eye-level camera drifts slightly backward; distant pedestrians stay in supporting roles. Choose this scene for an outdoor daytime setting, with a clear individual portrait rather than a photo of a plaza. The prompt requests recognizable identity, but the exact street layout and movement can vary.

Kitchen solo: a fixed phone viewpoint

The kitchen solo preset also requires one adult lead photo. White walls, a counter and open cubby shelves establish a bright kitchen and dining area with no other people. The requested sequence includes knee bounces, shoulder pulses, bent-arm sweeps, an arm flex and a hand heart before a small step toward the camera. A fixed phone viewpoint frames the head through the upper legs. Keep the shoulders and clothing readable in your photo; you do not need to photograph your own kitchen.

How to make your clip

  1. Select this template in the generator.
  2. Upload one portrait and choose vertical 9:16 or landscape 16:9.
  3. Generate a 15-second 720p clip for one credit, then review the result and audio.
Read the full walkthrough

Questions about this template

Does this edit my face into the example footage?

No. Your photo and a written scene prompt produce a new video. The example helps you choose the setting; it is not uploaded as a motion reference. Read how generation works.

Will the original Dame Hombro song play?

The request asks for newly generated instrumental and ambient audio, not the original song or a celebrity voice. See the trend and audio guide.

Can my friends join the same video?

For separate lead and supporting photo slots, choose the three-person template.

Is the dance copied at exact timestamps?

No. The written prompt requests shoulder bounces, claps, open palms and a turn toward the crowd. It does not assign exact timestamps or copy the example choreography frame by frame.

How is this different from the freestyle template?

The freestyle template requests a forward lean, alternating shoulder rolls and a hand beside the face in a medium shot. This main preset emphasizes claps and a brief turn toward the crowd in a waist-up view.

Do I need a nighttime photo?

No. Use a clear, authorized adult portrait with readable features. The red-lit entrance and crowd come from the scene prompt. Matching the sample lighting in your source photo is not a requirement.