Workflow name: Virtual Dress-up
[Workflow Introduction]
To open the workflow, you need to upload 2 images, one of which is the portrait model and the other is the clothes to be changed. Use Runninghub Captioner to reverse the prompt words + the prompt words you wrote. Combined with the FLUX large model, IC_TRY_ON_v3_e4LoRA model and the LoRA model raw image of the portrait, adjust the extended canvas and the picture composition to generate the effect of virtual dress-up.
[Usage scenario]
This workflow can directly display flat clothes through models, realizing the one-click dressing effect. Now the built-in parameter settings are complete. If you are interested, come and try it!
[Key nodes]
IC_TRY_ON_v3_e4 LoRA
[Model version]
FLUX
Model name: flux1-dev-fp8-Kijai.safetensors
[LoRA model]
Model name: IC_TRY_ON_v3_e4.safetensors
Model name: Shengshi Meiyan series.safetensors
ControlNetInpaintingAliMamaApply
Model: alimama FLUX.1-dev-Controlnet-Inpainting-Beta.safetensors
Strength: 0.9
Start time: 0
Introduction time: 1
[Custom sampler]
CFG: 3.5
Sampling method: euler
Scheduler: simple
Noise reduction: 1
Sign In