🚀 Flux2 Easy Swap: Head, Face or Outfit
About V1.1
Fixing Minor ıssues. Please upgrade your photobuf and mediapipe for "personmask" ( python_embeded\python.exe -m pip install --upgrade mediapipe protobuf google-cloud-videointelligence )
This workflow is optimized for the Flux model to perform natural and high-quality swaps of faces, heads, or outfits using PersonMaskUltra V2.
***HEAD OR FACE SWAP***
In the PersonMask Node
Img1 enable; body, clothe,accecories background
Img2 enable; face, hair
Prompt; It's best to leave this blank. Don't write anything. When you write something, the head size appears large and the colors may not match.
or
Prompt; "Replace the entire face and hair-style in image2 with the person image1. Make the face and body proportions perfect in the final image."
***OUTFIT CLOTHING SWAP***
In the PersonMask Node
Img1 enable; face,hair,body,background
Img2 enable; clothes
Prompt; "Style the person in image1, with every article of clothing in image2. In the final image, all the features of the clothes should be on the person in image1.
"
☕ Support My Work
Description
Fixing Minor ıssues. Please upgrade your photobuf and mediapipe for "personmask" ( python_embeded\python.exe -m pip install --upgrade mediapipe protobuf google-cloud-videointelligence )
FAQ
Comments (6)
I'm getting:
ValueError: operands could not be broadcast together with shapes (1348,933,1,4) (1348,933,4) (1348,933,4)
first of all; Please upgrade your photobuf and mediapipe for "personmask" ( python_embeded\python.exe -m pip install --upgrade mediapipe protobuf google-cloud-videointelligence )
Avoid Transparent Images: Ensure that your input images (IMG1 and IMG2) are standard RGB files (like JPG or non-transparent PNG). Transparent backgrounds or Alpha channels often trigger this dimension mismatch error during masking.
Force RGB Conversion: If you must use images with transparency, insert an "Image to RGB" or "Remove Alpha Channel" node immediately after your Load Image nodes. This ensures the tensor shapes remain consistent ($H, W, 3$) throughout the workflow.
Update Custom Nodes: Verify that your masquerade-nodes (Cut By Mask) and ComfyUI_LayerStyle_Advance (PersonMask) suites are fully updated. Also, ensure your environment is running numpy 1.26.4 for maximum stability with these specific nodes.
@ugurdoyduk341 Thanks for the reply. Everything was updated. I have fixed it with a short code snippet in the personmask v2 py file.
Gorgeous work. Thx for sharing.
Thanks!
Awesome! Had to disable Lora loader rgthree node cause it was throwing some errors, maybe cause there are no loras selected. Also used flux2k fp8 and flux2 vae from online platform...but bam, straight no fuss ♥