https://github.com/cubiq/ComfyUI_FaceAnalysis

Extension for ComfyUI to evaluate the similarity between two faces
https://github.com/cubiq/ComfyUI_FaceAnalysis

Last synced: 4 months ago
JSON representation

Extension for ComfyUI to evaluate the similarity between two faces

• Host: GitHub
• URL: https://github.com/cubiq/ComfyUI_FaceAnalysis
• Owner: cubiq
• License: apache-2.0
• Created: 2024-02-20T16:18:42.000Z (about 1 year ago)
• Default Branch: main
• Last Pushed: 2024-05-28T09:36:42.000Z (11 months ago)
• Last Synced: 2024-05-29T01:04:03.180Z (11 months ago)
• Language: Python
• Size: 167 KB
• Stars: 184
• Watchers: 2
• Forks: 19
• Open Issues: 5
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

• awesome-comfyui - **Face Analysis for ComfyUI**

README

        
# Face Analysis for ComfyUI

This extension uses [DLib](http://dlib.net/) or [InsightFace](https://github.com/deepinsight/insightface) to perform various operations on human faces.

The most obvious is to calculate the similarity between two faces. The best way to evaluate generated faces is to first send a batch of 3 reference images to the node and compare them to a forth reference (all actual pictures of the person). That will give you a baseline number that you can use to compare to generated images.

## Installation

You need to install either InsightFace or Dlib (or both).

For DLIB download [Shape Predictor](https://huggingface.co/matt3ounstable/dlib_predictor_recognition/resolve/main/shape_predictor_68_face_landmarks.dat?download=true), [Face Predictor 5 landmarks](https://huggingface.co/matt3ounstable/dlib_predictor_recognition/resolve/main/shape_predictor_5_face_landmarks.dat?download=true), [Face Predictor 81 landmarks](https://huggingface.co/matt3ounstable/dlib_predictor_recognition/resolve/main/shape_predictor_81_face_landmarks.dat?download=true) and the [Face Recognition](https://huggingface.co/matt3ounstable/dlib_predictor_recognition/resolve/main/dlib_face_recognition_resnet_model_v1.dat?download=true) models and place them into the `dlib` directory.

Precompiled Dlib for Windows can be found [here](https://github.com/z-mahmud22/Dlib_Windows_Python3.x).

![face analysis](./face_analysis.jpg)

The extension also supports [AuraFace](https://huggingface.co/fal/AuraFace-v1/tree/main) that is a free alternative to InsightFace. Download all the files and place them under `models/insightface/models/auraface/`