Upload one photo and drop yourself into a packed night-time baile charme under a Rio viaduct. Pick a template, keep the identity lines, and generate a vertical dance clip for TikTok, Reels, and Shorts.
Updated October 20, 2026

Pick the job you need to finish. The generator, photo-to-video, prompt, trend, and CapCut pages all use the same six templates and the same uploaded photo.
Six original baile charme templates built around the trend’s visual language: the packed viaduct crowd, the side-step, the friend, the couple, the older dancer, black-and-white film, and the 80s action remix.






Use a front-facing photo with even light and no sunglasses. Show your hair, shoulders, and outfit. For a couple or a friend, use one photo with both people or compose two portraits first.
The prompt protects the face, skin tone, body proportions, and the number of people. Change the outfit, the film look, or the camera line — not the person.
Generate the still first. When the face and outfit look right, send that image to Veo 3.1 Fast image-to-video for a 9:16 clip of up to 10 seconds.
This is the base baile charme recipe. Every template page adds its own outfit, scene, and motion.
Open the generator, upload your photo, and keep the identity and anatomy lines intact. The still prompt is written for an image-editing model; the motion prompt is written for image-to-video.
Use the uploaded photo as the primary reference. Preserve the subject's recognizable face, facial proportions, skin tone, approximate age, body proportions, and natural anatomy. Keep the same number of people as the uploaded photo. Place the subject in a packed outdoor baile charme at night under a concrete viaduct, with warm string lights, a dancing crowd, and a handheld phone framing. Photorealistic editorial photography, realistic skin texture and fabric, warm night-party lighting, handheld phone perspective, shallow depth of field. Avoid altered identity, extra limbs, duplicated fingers, distorted anatomy, text, logos, watermarks, and celebrity likenesses.
Start with an identity-first image model, then animate with a video model that holds a face in a moving crowd. The one-click routes below use models enabled in the generator.
It is a short AI video that places a person in a crowded baile charme, the Black Brazilian partner-dance party that grew around the Viaduto de Madureira in Rio de Janeiro. The clip shows the side-to-side charme step, a handheld phone angle, and a crowd under warm lights. Vidle rebuilds that look from one photo.
No. CapCut templates swap a face into a fixed video. Vidle generates a new clip from your photo and a prompt, so the face, outfit, and scene can change. You can still import the MP4 into CapCut for captions or the trending sound.
No. You need one clear photo. The prompt describes the camera, the step, and the crowd; the video model animates the still. A front-facing photo with visible shoulders gives the most stable result.
Generate the still first. When the face and outfit look right, send that image to Veo 3.1 Fast image-to-video for a 9:16 clip of up to 10 seconds.
Independent AI video tool. Not affiliated with, endorsed by, or sponsored by the Baile Charme de Madureira, its organisers, TikTok, CapCut, or any artist. Only upload photos of people who gave you permission, all 18 or older. Do not upload celebrities or public figures.