Unlock Simple Face Recognition in Python: Install, Use, and Explore Features

Face Recognition Package

This is the world's simplest face recognition library. You can use it via Python import or a command‑line tool to manage and identify faces.

The package leverages dlib's state‑of‑the‑art deep learning algorithm, achieving 99.38% accuracy on the Labeled Faces in the Wild benchmark.

Features

Detect faces in images

Find all faces in a picture.

Locate facial landmarks

Obtain positions and contours of eyes, nose, mouth, and chin.

These landmarks enable many applications, such as beauty filters.

Identify faces in images

Recognize who is in a picture.

You can also perform real‑time face recognition.

Example:

https://github.com/ageitgey/face_recognition/blob/master/examples/facerec_from_webcam_faster.py

Installation

Environment requirements

Python 3.3+ or Python 2.7

macOS or Linux (Windows is unofficial but may work)

Installation steps

On macOS/Linux

First ensure dlib and its Python bindings are installed (see

https://gist.github.com/ageitgey/629d75c1baac34dfa5ca2a1928a7aeaf

).

Then install the package with pip: pip install face_recognition If you encounter issues, try the pre‑configured virtual machine (

https://medium.com/@ageitgey/try-deep-learning-in-python-now-with-a-fully-pre-configured-vm-1d97d4c3e9b

).

On Raspberry Pi 2+

See

https://gist.github.com/ageitgey/1ac8dbe8572f3f533df6269dab35df65

.

On Windows

Community guide:

https://github.com/ageitgey/face_recognition/issues/175#issue-257710508

.

Usage

Command‑line interface

After installation, the face_recognition command is available. Provide a folder of known faces (one image per person, named with the person’s name) and a folder of unknown images, then run: face_recognition /path/to/known /path/to/unknown The output lists each detected face as <image_name>,<person_name>.

Python module

Import the face_recognition module and follow the documentation ( https://face-recognition.readthedocs.io).

Automatic face detection

Example:

https://github.com/ageitgey/face_recognition/blob/master/examples/find_faces_in_picture.py

You can replace the underlying deep‑learning model; GPU acceleration via NVIDIA CUDA is recommended for better performance.

GPU example:

https://github.com/ageitgey/face_recognition/blob/master/examples/find_faces_in_picture_cnn.py

For batch processing with many images and GPUs, see

https://github.com/ageitgey/face_recognition/blob/master/examples/find_faces_in_batches.py

.

Facial landmark detection

Example:

https://github.com/ageitgey/face_recognition/blob/master/examples/find_facial_features_in_picture.py

Identify who is in a picture

Example:

https://github.com/ageitgey/face_recognition/blob/master/examples/recognize_faces_in_pictures.py

More Resources

Visit the project repository for additional documentation and examples:

https://github.com/ageitgey/face_recognition