Magic Portrait
5.0
0 reviewsDescription
Magic Portrait
描述:快速生成肖像照片,支持风格参考/自定义 两种模式
模式选择
- 风格参考模式:开启`Use Img Style Reference`,使用ipadapter进行风格指引
- 自定义模式:关闭`Use Img Style Reference`,根据提示词生成风格参考图,使用instantid生成人像一致的图片
概念滑块
- 胖瘦: 调节chubby,建议权重-3~3
- 笑容:调节smile,建议权重-3~3
- 年龄: 调节age,建议权重-3~3
Description: Quickly generates portrait photos, supporting two modes - Style Reference and Customization.
Mode Select
- Style Reference Mode: Turn on "Use Img Style Reference" and use ipadapter for style guidance.
- Customization Mode: Turn off "Use Img Style Reference", generate a style reference image based on prompt words, and use instantid to create a consistent portrait image.
Concept Slider
- Body Size: Adjust "chubby", recommended weight -3 to 3
- Smile: Adjust "smile", recommended weight -3 to 3
- Age: Adjust "age", recommended weight -3 to 3
Support
You can buy me a coffee here: https://ko-fi.com/alansun
Thanks to your support!
Discussion
(No comments yet)
Loading...
Resources (3)
Reviews
No reviews yet
Versions (1)
- latest (5 months ago)
Node Details
Primitive Nodes (19)
Anything Everywhere
Fast Groups Muter (rgthree)
GetNode
Note
PrimitiveNode
Reroute
SetNode
Custom Nodes (36)
CLIPSeg
- CLIPSeg
- CR Apply LoRA Stack
- CR LoRA Stack
- CR Text Concatenate
ComfyUI
- GrowMask
- EmptyLatentImage
- CLIPVisionLoader
- CheckpointLoaderSimple
- KSampler
- CLIPTextEncode
- VAEDecode
- PreviewImage
- LoadImage
- SaveImage
- InstantIDFaceAnalysis
- InstantIDModelLoader
- ApplyInstantIDAdvanced
- WD14Tagger|pysssss
- IPAdapterNoise
- IPAdapterUnifiedLoader
- IPAdapterAdvanced
- ControlNetLoaderAdvanced
- If ANY execute A else B
- Bool
- TextInput_
- ShowText|pysssss
- ConstrainImage|pysssss
- String Literal
- Bounded Image Crop with Mask
- Image Input Switch
Model Details
Checkpoints (1)
sdxl\juggernautXL_v9Rdphoto2Lightning.safetensors
LoRAs (3)
xl滑块\age.pt
xl滑块\chubby.pt
xl滑块\smiling.pt