Image Generation AI with ComfyUI: Face Detailer Edition

Progress 12 / 15
Table of Contents

Face Detailer is one of techniques when generating image with AI, and can repair only face part which tends to collapse in image generation. (Inpainting masking face)

In this article I introduce method to try Face Detailer easily with ComfyUI.

Work Procedure

1. Introduction of ComfyUI

Introduce ComfyUI. Those already introduced please update ComfyUI to latest version too.

【Stable Diffusion】ComfyUIを使って画像生成AIで遊んでみよう【導入編】

>-

blog.otama-playground.com

2. Installation of Extension

Install following extension.

GitHub - ltdrdata/ComfyUI-Impact-Pack: Custom nodes pack for ComfyUI This custom node helps to conveniently enhance images through Detector, Detailer, Upscaler, Pipe, and more.

Custom nodes pack for ComfyUI This custom node helps to conveniently enhance images through Detector, Detailer, Upscaler, Pipe, and more. - ltdrdata/ComfyUI-Impact-Pack

github.com

3. Import Workflow

Import and use following workflow. Extending default workflow, adding FaceDetailer to flow after generation.

face_detailer.json
drive.google.com

4. Generate

Adjust prompt and model and generate.

Generation Result

Using SD1.5 model where face tends to collapse easily for easy understanding.

Before FaceDetailer
Before FaceDetailer
After FaceDetailer
After FaceDetailer

Conclusion

This time I introduced method to regenerate only face using FaceDetailer in ComfyUI.

Since it often becomes able to generate well if incorporating into workflow when feeling face easily collapses, I think it is good to keep as option. In my case when using SD1.5 model, or generating video with AnimateDiff, I use it often.