How I Cloned Myself Into a Two-Man Band, Then Turned Us Into Angine de Poitrine

How I Cloned Myself Into a Two-Man Band, Then Turned Us Into Angine de Poitrine

I recorded myself playing drums and guitar, merged the two takes so it looks like two of me jamming in the same room, then used AI to turn both of me into the guitarist and drummer from Angine de Poitrine. The movement is all mine: my real performance drives the AI version. Here's how to do it yourself.

WHAT YOU NEED

  • A phone or camera on a tripod
  • A way to record your song (any DAW, or just your phone)
  • CapCut for editing
  • A Higgsfield account for the AI image and the motion transfer

STEP 1: Record the song first
Record the full song with every instrument before you film anything. You'll play along to this track on camera, so every take lines up with the same audio.

STEP 2: Film part one
Lock the camera on a tripod, horizontal (16:9). It must not move at all between takes. Play the song through a speaker or earbuds and film yourself on the first instrument (drums for me). Stay on your side of the frame.

STEP 3: Film part two
Don't touch the camera or change the lights. Move to the other side of the frame and film yourself playing the second instrument (guitar for me) to the same track. Keep your two positions from overlapping. A clean empty gap between you makes the merge much easier.

STEP 4: Merge the two takes into one "clone" video
In CapCut, put take one on the main track and take two on an overlay track directly above it. Sync both to the song, using a clear hit or count-in to line them up. Add a mask (split or rectangle) to the top clip so it only shows your side of the frame. Feather the mask edge a little so the seam disappears. Export it. You now have one video of two of you playing in the same room.

STEP 5: Grab the first frame
Move the playhead to 0:00 on the merged video and export that frame as a full-resolution image. This is your starting image for the AI. Taking it from the very first frame keeps the image and the video perfectly lined up for motion transfer.

STEP 6: Transform the still in Higgsfield
Upload the still to Higgsfield and pick an image-editing model. I used GPT Image 2.5 at high quality, 2K, 16:9. If you have a reference photo of the look you're going for, upload that too. Start the prompt by locking everything that has to stay the same, then list the changes. Here's a template:

"Edit this photo. Keep the exact same camera angle, framing and composition. Keep both people in exactly the same positions and poses, with their hands in the same places on their instruments. Change the person on the left into [describe the drummer look] and the person on the right into [describe the guitarist look]. Keep the drum kit and guitar in the same positions. Change the room into [describe the setting]. Photorealistic, natural hands with five fingers."

STEP 7: Check the image before animating
Compare it side by side with your original frame. Is the camera angle the same? Are both people in the same spots? Are the hands on the sticks and the guitar neck where yours were? Are the instruments still realistic? If something's off, fix that one thing in the prompt and generate one image at a time. Don't burn credits rolling batches blindly.

STEP 8: Transfer your motion onto the new image
In Higgsfield, open Genjutsu Motion Control. Add your new AI image as the image reference and your merged clone video as the video reference, and set it to 1080p. Prompt:

"Transfer the exact body, arm, hand and finger motion from the reference video onto both people in the image. Keep their looks, instruments and the background exactly as in the image. Keep the camera static and matching the image framing. Photorealistic."

STEP 9: Review and add your song back
Watch it all the way through, especially the drumstick hits and the guitar fretting hand, since fast hand movement is the hardest part. Download the MP4, bring it into CapCut, and lay your original song over it so the audio is crisp.

STEP 10: Show the process and post
The before-and-after is what makes people stop scrolling. Try a vertical split screen with your raw takes, the merged clone video, and the final AI version stacked on top of each other. Export at 1080x1920 and post.

TIPS THAT SAVE CREDITS

  • The tripod is everything. If the camera moves even slightly between takes, the merge and the motion transfer both break.
  • Leave space between your two positions. Overlapping bodies are hard to mask and confuse the AI.
  • Always start your image prompt by locking the camera and poses. Otherwise the AI redraws the whole scene from a new angle.
  • Think about how the new costumes and props would really look from your camera angle so nothing ends up blocking your hands.

Shoutout to Angine de Poitrine for the inspiration. Go check them out.

Back to blog