faceai
Beginner-level Python demos for face, video, and text detection built on OpenCV and Dlib.
- Stars
- 11.1k
- Forks
- 2.5k
- Licence
- MIT
- Language
- Python
- Last update
- 6.6 years ago
- Type
- Learning resource
No commits in over a year — check for forks or alternatives before adopting.
What it is
faceai is a Python project of tutorials and example code for detecting and recognizing faces in images and video. The README also covers features like drawing face outlines, adding hats and makeup to faces, gender and emotion recognition, OCR text recognition, and image inpainting. It is written in Chinese and is described by its author as an introductory project.
- Run face detection on a still image or a video file using OpenCV or Dlib
- Draw a face outline and the 68 facial landmarks on a photo
- Recognize emotion or gender from a face using the included Keras models
- Try OCR text recognition with Tesseract
Great for
- Students learning basic computer vision with Python
- Developers who want working examples of OpenCV and Dlib face detection
- Readers who can follow Chinese-language tutorials
Not ideal for
- Production systems, since the README documents a 2018-era setup and the repository's last push was in April 2020
- Readers who cannot read Chinese, since the main docs are in Chinese (an English README is linked but not included here)
- People who need the setup to run on current Python or library versions, since the README lists Python 3.6.4 and TensorFlow 1.8.0
How to get it
- 1
Clone and build from source
You'll need: Python 3.6.4 (version listed in README) · OpenCV 3.4.1 (version listed in README) · Dlib 19.8.1 (version listed in README) · Tesseract OCR 4.0.0-beta.1 for OCR features (version listed in README)
Clone the repository
git clone https://github.com/vipstone/faceai - 2
Run it
- Install the listed Python and library versions, following doc/settingup.md for OpenCV setup
- Open the tutorial for the feature you want, such as doc/detectionOpenCV.md for image face detection
- Run the example script described in that tutorial
- 3
Get the most out of it
- Follow the linked tutorials in order, starting with the OpenCV setup guide
- Use the Dlib tutorials to compare detection methods against the OpenCV versions
- Try the Tesseract OCR guide if you want text recognition
Downloads
Strengths
- Covers many basic computer vision tasks in one project
- Tutorials are organized by feature and linked from the README
- Uses widely known libraries such as OpenCV, Dlib, and Keras
Trade-offs
- Dependencies are pinned to old versions (Python 3.6.4, TensorFlow 1.8.0), which may be hard to install today
- Several features are marked unfinished, including eye tracking and face swapping
- Last push was April 2020, so it may not reflect current library APIs
How it compares
| Project | Stars | Type | Level | Updated | Best for |
|---|---|---|---|---|---|
| faceai (this) | 11.1k | Learning resource | beginner | 6.6 years ago | Students learning basic computer vision with Python |
| face_recognition | 56.9k | Library | intermediate | 4 months ago | Python developers prototyping photo sorting or face matching |
| netron | 33.6k | App | beginner | 5 days ago | Engineers who need to check a model's layers and tensor connections before deploying it |
| wandb | 11.3k | Library | beginner | 4 days ago | Machine learning engineers who want to compare training runs over time |
| TensorFlow-Course | 16.3k | Learning resource | beginner | 3.9 years ago | People who want to learn TensorFlow by running example notebooks |
| nlp-tutorial | 14.9k | Learning resource | intermediate | 2.7 years ago | Students who want to see how classic NLP models work in code |
Stats synced from GitHub. The assessment above was drafted with AI from the project's README on 2026-10-29 and may miss recent changes — check the README before relying on a command. MIT licence: free to use, modify and distribute, including commercially, as long as you keep the copyright notice.