Revolutionize Print Pattern Design with AI-Powered Workflow
1. Workflow Overview
.png)
This professional workflow redesigns print patterns through:
Original pattern analysis (Florence-2)
AI regeneration (Stable Diffusion + Flux)
Precision masking (Segment Anything)
Garment compositing
Core Models:
F.1-dev-fp8_1
: FP8-optimized main modelFlorence-2-base
: Visual understandingsam_vit_h
: Segmentation model
2. Node Breakdown
Key Nodes:
Florence2Run
Auto-generates prompts from patterns
SegmentAnythingUltra V2
Hair-accurate masking
Requires
VITMatte
algorithm
ImageBlendAdvance
90% opacity control
Lanczos resampling
Dependencies:
ComfyUI-LayerUtility
pluginDownload
GroundingDINO_SwinT
(694MB)
3. Workflow Stages
Pre-Processing
Upscale input pattern 4x with ESRGAN
Load shirt template (1024x1280px)
Generation
FluxGuidance (CFG=3.5)
20-step Euler sampling
Compositing
Light/shadow matching
Side-by-side comparison
4. I/O Specification
Inputs:
Source print:
印花.png
Garment:
white_shirt1.png
Outputs:
Real-time preview
Interactive comparer
5. Critical Notes
Hardware
FP8 requires RTX 30/40 series
Recommended VRAM: ≥12GB
Model Paths
Florence-2 auto-downloads to:
ComfyUI/models/florence2/
Manually install SAM models:
ComfyUI/models/sam/
Troubleshooting
Rough edges:
Adjust mask_threshold (default:0.3)
Poor garment fit:
Change blend mode to "multiply"