Workflow Name: E-commerce Clothing Custom Pose Consistency Model Generation
【Workflow Introduction】
Workflow operation process: First, open the workflow and upload one image on the far left side of the screen. Use the ControlNet skeletal model to generate consistent human characters. Then, split the image into 4 poses and perform segmentation and overlay processing using SegmentAnythingUltra_V2. Next, use the flux1 redux dev model for image style transfer, and then use the flux1 fill dev model to redraw the masked area. Finally, the sampler generates the output image. This workflow has many nodes and is divided into 2 parts. At this stage, it focuses on multi-pose model clothing transfer while ensuring consistency.
Current workflow status and application suggestions: The current image is somewhat blurry. It is recommended to run a high-definition portrait restoration workflow, with the link available in the workflow note introduction.

【Application Scenarios】
This workflow can accurately transfer specified clothing to the generated model while ensuring character consistency, making it highly suitable for the needs of personnel in the e-commerce clothing field.

【Key Nodes】
FLUX redux dev

【Model Version】
FLUX
Model Name: flux1 redux dev
【LoRA Model】

Model Name: FLUX.1 Turbo Alpha

【ControlNet Application】

Start Time: 0
End Time: 0.6
Strength: 0.4

【K Sampler】
CFG: 4.2
Sampling Method: deis
Scheduler: beta
Denoising: 1