
Use one clearly visible adult
- An unobstructed face and visible hair in even lighting.
- The torso and outfit you want to use; a full-body photo helps define trousers and shoes.
- One person, facing the camera or at a slight angle.
Your photo in the Choosin’ Texas emo rooftop cover.Put yourself in the Choosin’ Texas rooftop cover. Your face, hair and outfit join the exaggerated emo performance and reference recording.
The Choosin’ Texas emo remix turns the song into an exaggerated rooftop performance. Make your own visual take on that cover with one adult in the performer's role, using a short excerpt from the opening.
Your photo guides the performer's face, hair and visible clothes. The reference supplies the nighttime parking-garage rooftop, low cream-colored wall, city lights, framing and on-screen captions.
Watch @kobbapot’s original TikTokSee the original rooftop performance here, and the actual 15-second AI result with its input photo in How to.
Both clips start with the opening leg lift, just after the brief camera lift from the ground. Choose 5 seconds or continue the performance to 15 seconds, with the matching reference audio.
Upload a clear photo of yourself or a consenting adult. That person takes the single performer role, with the photo guiding face, hair and clothing.
The nighttime rooftop, low cream-colored wall and city lights stay with the reference. The background in your photo does not become the video scene.
The opening includes a raised-leg gesture; the 15-second excerpt continues with wall-side crouching and expressive hand movements. The reference guides the timing, while generated details can vary.
One adult photo takes the rooftop performer role.
Choose one clear adult photo with an unobstructed face, visible hair and the outfit you want the performer to wear.
Preview the reference, choose 5 or 15 seconds and 480P or 720P, then check the total credits before submitting.
Watch the finished video before downloading it from your result or History. Check face, clothing, hands and timing in motion.
Choose a photo that clearly shows the person and outfit you want in the rooftop cover.


A single performer role connects your appearance to this recognizable performance.
Use a photo that shows the look you want to bring into the scene. Fine details may change during movement.
The reference supplies the background and framing, keeping the remake focused on the same rooftop cover.
The gestures follow the selected clip, and the soundtrack uses its recording. Your photo does not create a new singing voice.
Check the selected settings and current total before each generation.
Choose a 5-second or 15-second opening excerpt at 480P or 720P. The How to example is an actual 15-second, 480P result.
The generation button shows the current credit total for the selected length and resolution.
This rooftop cover and Motion Transfer use the image and reference differently.
Explore more scenes that put your photo into a reference performance.
It makes a visual remake of the Choosin’ Texas emo rooftop cover, with your uploaded adult photo in the performer role. The nighttime parking-garage rooftop, expressive movement and reference recording define this template. Each option uses a short opening excerpt.
Use one photo of yourself or a consenting adult. This template is designed for a single adult performer; pets, children and multiple-person uploads are outside its supported input.
Your photo guides face, hair and visible clothing. Show the torso and outfit clearly, and include the full body if trousers and shoes matter. Small patterns, accessories and clothing details can vary in the result.
The scene is the fixed rooftop from the reference. The uploaded image supplies the person's appearance, not a replacement background. Use Motion Transfer if you want to animate a person within your photo's scene.
The soundtrack uses the reference recording for the selected excerpt. Uploading a photo does not clone your voice or create a new vocal performance. Review the mouth movements and audio timing in the finished video.
Choose 5 or 15 seconds at 480P or 720P. Both clips begin with the opening leg lift, after the original video’s brief camera lift from the ground. These excerpts do not include the later kneeling ending. The How to shows an actual 15-second, 480P generation.
The reference guides the performance, but exact gestures, hand shapes, facial details and mouth timing can vary. Choose the excerpt for the action it actually shows and review the finished video before sharing.
Use History to follow the task status and open a completed result. Failed tasks follow the site's existing credit-return handling. Check the status and balance before submitting another paid attempt; a completed video can still have visual flaws.
Review the result, identify it as an AI-edited remake and credit the original reference where appropriate. Use a photo you have permission to upload, and check the music and platform rules that apply to your intended use. The generated clip is not a recording of you actually singing on that rooftop.