My stuff

  • My Workflows

  • Liked Workflows

  • Following Workflows

Go to OpenArt main site
Upload workflow

One-Click Change Hairstyle

5.0

0 reviews
29
6.1K
1.6K
3
Description

该工作流基于SD1.5版本创作(也是本人第一个工作流)

可以根据用户提供的头发模板图片给人物切换发型,支持短发切换到长发,长发切换至短发发丝消除效果不佳。

使用的大模型包括:Dreamshaper_8/Deamshaper_8Inpainting

使用到的lora包括:BaldfierW2,add_detail

其他:openpose,ipadapter

详情下载工作流查看参数。


残留问题:

q1:去除头发导致头骨变大

q2:头发与图片的色差太大

q3:头发很假

q4:耳朵处出现修补空白

q5:   脸部遮挡无法修复

q6:   刘海检查难题

工作流残留问题有优化思路,大家可以一起讨论,感谢社区,感谢开源大佬的贡献。


Hair change workflow, the workflow based on SD1.5 version of the creation (also my first workflow)

You can switch hairstyles for characters according to the hair template image provided by the user, perfect support for short hair switch to long hair, long hair switch to short hair hair elimination effect is not good.

Use of large models include: Dreamshaper_8/Deamshaper_8Inpainting

The lora used include: BaldfierW2,add_detail

Others: openpose, ipadapter

For details download the workflow to see the parameters.


Residual issues:

q1: Removing hair causes skull to become larger

q2: Hair is too different in colour from the picture

q3: Hair is very fake

q4: Repair gaps at the ears

q5: Face masking cannot be repaired

q6: Bangs checking difficulties

There are optimization ideas for residual workflow issues, and we can discuss them together. Thank you to the community and open source experts for their contributions.

Discussion

(No comments yet)

Loading...

Author

3
18.0K
293
54.9K

No reviews yet

  • - latest (8 months ago)

  • - v20250206-070805

Primitive Nodes (8)

Fast Groups Bypasser (rgthree) (1)

Florence2ModelLoader (1)

Florence2Run (1)

Image Comparer (rgthree) (2)

Note (2)

Reroute (1)

Custom Nodes (47)

ComfyUI

  • - VAEDecode (2)

  • - ControlNetApplyAdvanced (2)

  • - ControlNetLoader (2)

  • - InpaintModelConditioning (1)

  • - KSampler (2)

  • - PreviewImage (6)

  • - LoraLoader (2)

  • - CLIPTextEncode (4)

  • - ImageUpscaleWithModel (1)

  • - CheckpointLoaderSimple (2)

  • - LoadImage (2)

  • - SaveImage (1)

  • - GrowMask (3)

  • - VAEEncode (1)

  • - UpscaleModelLoader (1)

ComfyUI Layer Style

  • - LayerMask: PersonMaskUltra V2 (2)

  • - LayerUtility: ImageScaleByAspectRatio V2 (2)

  • - LayerUtility: ImageBlend V2 (3)

  • - IPAdapterAdvanced (1)

  • - IPAdapterUnifiedLoader (1)

  • - FaceAnalysisModels (2)

  • - FaceSegmentation (2)

Checkpoints (2)

SD1.5\dreamshaper_8.safetensors

SD1.5\dreamshaper_8Inpainting.safetensors

LoRAs (2)

SD1.5\调整\BaldifierW2.safetensors

SD1.5\调整\add_detail.safetensors