https://github.com/anandbaburajan/multiple-face-recognition
Train multiple images per person and get started with recognizing faces with the face_recognition API
https://github.com/anandbaburajan/multiple-face-recognition
face-detection face-recognition python svc
Last synced: 11 days ago
JSON representation
Train multiple images per person and get started with recognizing faces with the face_recognition API
- Host: GitHub
- URL: https://github.com/anandbaburajan/multiple-face-recognition
- Owner: anandbaburajan
- License: mit
- Created: 2019-07-25T08:10:11.000Z (almost 6 years ago)
- Default Branch: master
- Last Pushed: 2019-12-30T10:00:47.000Z (over 5 years ago)
- Last Synced: 2025-03-28T05:02:58.423Z (28 days ago)
- Topics: face-detection, face-recognition, python, svc
- Language: Python
- Homepage:
- Size: 2.81 MB
- Stars: 6
- Watchers: 0
- Forks: 2
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
Multiple Face Recognition
=================
Train multiple images per person then recognize known faces in an image using a SVC in Python. One can easily use this to develop apps such as face-based attendance systems.
This program is based on ageitgey's [face_recognition](https://github.com/ageitgey/face_recognition) api for Python and [dlib](https://github.com/davisking/dlib). A support vector classifier (SVC) with scikit-learn is trained on the face encodings from all the known faces in the training directory. It then recognizes the faces found in a test_image. Please note that it will produce meaningless results on very small datasets.
Installation
------------
```bash
pip install -r requirements.txt
```
Usage
-----
```bash
Usage:
face_recognize.py -d -i
Options:
-h, --help Show this help
-d, --train_dir= Directory with images for training
-i, --test_image= Test image
```
Training directory structure
-----
```
/
/
.jpg
.jpg
.
.
.jpg
/
.jpg
.jpg
.
.
.jpg
.
.
/
.jpg
.jpg
.
.
.jpg
```