InstantID Face Regeneration
created a year ago
inpainting
img2img
face
controlnet
20 nodes

2.7k

484

Outputs
example image output 0 for workflow.
example image output 1 for workflow.
example image output 2 for workflow.
example image output optimized_comfy_workflows_user_uploads/780bfaed-126f-4081-a1c2-eb8d5bcedae7/assets/r5.webp for workflow.
Description

Use the key point information from the original image, CLIPSeg for automatic inpaint area detection and differential diffusion for inpainting over the face.

Built-in nodes
DifferentialDiffusion
Custom nodes
CLIPSeg Model Loader
InstantIDFaceAnalysis
InstantIDModelLoader
ControlNetLoader
CLIPSeg Masking
VAEEncode
SetLatentNoiseMask
GrowMaskWithBlur
PreviewImage
LoadImage
ApplyInstantIDAdvanced
ConditioningZeroOut
CheckpointLoaderSimple
CLIPTextEncode
KSampler
SaveImage
VAEDecode
0
0
0
0