https://github.com/dreamsoftin/facetool

Simple Python script to crop and align faces
https://github.com/dreamsoftin/facetool

Last synced: 2 months ago
JSON representation

Simple Python script to crop and align faces

• Host: GitHub
• URL: https://github.com/dreamsoftin/facetool
• Owner: dreamsoftin
• License: mit
• Created: 2021-06-13T12:48:31.000Z (about 4 years ago)
• Default Branch: master
• Last Pushed: 2021-06-17T07:41:56.000Z (about 4 years ago)
• Last Synced: 2023-03-02T15:09:25.147Z (over 2 years ago)
• Language: Python
• Size: 1.35 MB
• Stars: 4
• Watchers: 2
• Forks: 2
• Open Issues: 1
• Metadata Files:
  • Readme: README.md
  • License: LICENSE.txt

Awesome Lists containing this project

README

        
# FaceTool

A simple python package to crop and align faces. Uses Mediapipe for face detection

|Input |Output|

--- | --- |

|![before](assets/before.jpg "before")|![after](assets/after.jpg "before")|

|![before](assets/before2.jpg "before")|![after](assets/after2.jpg "before")|

---

|Output |ZoomFactor 1| ZoomFactor 4| Offset -200|

--- | --- | --- | ----

|![before](assets/after3.jpg "before")|![after_zoom_1](assets/after_zoom_1.jpg "after_zoom_1")|![after_zoom_4](assets/after_zoom_4.jpg "after_zoom_4")|![after_offset_200](assets/after_offset_200.jpg "after_offset_200")

---

## Requirements

* Python 3.6

* MediaPipe

* OpenCV

install all dependencies

```bash

pip install -r requirements.txt

```

---

## Usage

```PYTHON

facetool = FaceTool(debug=False)

#Crop and Align, Returns 3 Parameters (face_count, faces array)

face_count, faces, message = facetool.crop_and_align(image, size=512, zoom_factor=2, offset_y=-10)

if face_count > 0:

    # face are sorted by area

    # takes the face with largest area.

    cv2.imshow("cropped_image_0", faces[0].image)

```

---

Acknowledgements

----------------

The alignment code is partially Based on the tutorial by [datahacker.rs](http://datahacker.rs/010-how-to-align-faces-with-opencv-in-python/)

This repository contains some stock photos from [Pexels].

[Pexels]: https://www.pexels.com/