Choose the template
Dame Hombro includes Baby, Solo, Duo and Trio. The Bulin 47 tab opens the existing freestyle template. Each card shows how many photos it needs.
Turn your photos into a Dame Hombro AI video. Pick the Baby, Solo, Duo or Trio shoulder-dance template, or open the Bulin 47 tab for the original freestyle. Each template has its own prepared scene, camera and timing.
Pick a trend tab, choose a prepared template and upload the matching number of photos. There is no prompt to write and no timeline to edit.
Dame Hombro includes Baby, Solo, Duo and Trio. The Bulin 47 tab opens the existing freestyle template. Each card shows how many photos it needs.
Use one clear photo per role. The form changes from one compact slot to two or three when the selected template needs more performers.
Choose 480P, 720P or 1080P, confirm the frame and credit cost, then generate. Failed tasks return unused credits automatically.
Every template keeps its own location, camera, choreography and timing. Your photos fill the identity-bearing roles while the scene and action remain the prepared performance.
Baby, Solo, Duo and Trio each have their own video reference and their own role layout. Switching templates changes both the preview and the upload slots.
The Bulin tab reuses the existing local Bulin template and prompt, including the single centre-performer role and the original street-freestyle staging.
Auto follows the selected template. Choose 9:16 for vertical feeds or 1:1 for a square crop when that shape fits your post.
The output ratio sets the canvas while Baby supplies the motion, camera and timing.
Baby leads the page, with Solo, Duo, Trio and the Bulin 47 freestyle available as separate prepared clips.
Each template has its own private scene and performance instructions. Public pages describe the outcome and the photo requirements; they do not expose internal generation instructions.
It is a multi-template tool with a clear cast count, not a prompt box that asks you to invent a scene.
The Dame Hombro shoulder dance is easy to recognise because of its fixed cast positions and simple shoulder-led movement. The Baby clip takes that energy into a home hallway. Solo gives one performer the full frame. Duo adds a second role, and Trio returns to the grey roller shutter with three performers and a clear centre position.
The Bulin 47 freestyle is a different performance with its own venue, crowd, camera and timing. Reusing the existing Bulin template and prompt keeps that tab faithful to the original local Bulin project instead of flattening both trends into one generic dance scene.
Every template accepts a different number of photos because every source video has a different number of identity-bearing performers. The generator keeps those roles explicit, which prevents an extra person from appearing and keeps the upload form compact.
Avoid group photos where the intended person is ambiguous. Baby should be alone, Solo should show one performer, Duo needs two separate people and Trio needs three.
Shoulder movement, arm gestures, steps, camera drift and the final timing come from the selected clip. You do not need to describe the action in a prompt.
The Dame Hombro tab contains Baby, Solo, Duo and Trio. The Bulin 47 tab contains the existing freestyle template and its one-photo role.
480P is the low-cost option, 720P is the normal social choice and 1080P is for a larger screen. The required credits appear before submission.
Give each role a readable, recent photo with one clear face.
Use even light and keep the face unobstructed. Sunglasses and heavy filters reduce the information available for the finished performance.
For Trio, upload the left, centre and right photos in the order shown. For Duo, keep the left and right order visible in the form.
Only upload a baby or minor's photo if you are their parent or legal guardian and have permission to use it. The Baby template is intended for personal family clips.
Dame Hombro AI is a browser-based video generator built around the shoulder-dance trend and the Bulin 47 freestyle. The Dame Hombro tab includes four prepared templates: the Dame Hombro Baby clip in a bright home hallway, a Solo performance, a Duo in a hallway and a Dame Hombro trio dance in front of a grey roller shutter. The Bulin 47 tab reuses the original freestyle scene, the centre-performer role and the same prepared performance guidance, so it behaves like the existing Bulin 47 tool rather than an imitation.
A Dame Hombro AI video starts with a photo for each role. Baby and Solo need one photo, Duo needs two and Trio needs three. The generator keeps the location, camera, choreography, cast positions and timing from the selected template. Uploaded backgrounds and photo poses do not enter the result. When you switch templates, the upload form changes to the correct number of compact photo slots, so the three-person mode stays readable instead of stretching the tool across the page. This is also how the Bulin 47 AI video flow works: one photo fills the centre performer role while the crowd, venue and delivery stay fixed.
Searches for AI Dame Hombro video, Dame Hombro generator, Dame Hombro dance and the Bulin 47 AI trend often describe the same pair of trends. The Bulin 47 AI video generator here reuses the Bulin 47 freestyle AI setup from the existing local project. This page brings both trends together without asking you to write a prompt or edit a timeline. Dame Hombro AI Video Generator is built to follow the original: the original camera, choreography, cast positions and timing. Every finished render includes the same prepared audio bed across all five templates, so the downloaded MP4 is published with sound. Output is available in 480P, 720P and 1080P, with one-time credit packs and automatic credit refunds when a task fails before delivery.
Understand the trend, compare the templates and walk through the one-to-three-photo workflow.
Template counts, photo requirements, Bulin 47, pricing and what the finished clip includes.
Need help with credits, a failed task or a paid order? Contact support.
Choose a template, upload the matching photos and let the prepared performance carry the rest.