Face Recognition is the world's simplest face recognition library. It allows you to recognize and manipulate faces from Python or from the command line using dlib's (a C++ toolkit containing machine learning algorithms and tools) state-of-the-art face recognition built with deep learning.

Face Recognition is highly accurate and is able to do a number of things. It can find faces in pictures, manipulate facial features in pictures, identify faces in pictures, and do face recognition on a folder of images from the command line. It could even do real-time face recognition and blur faces on videos when used with other Python libraries.

Features

  • Finding faces in an image (directly or via deep learning)
  • Finding faces in a large number of images w/ GPU (through deep learning)
  • Blurring all the faces in a live video from your webcam (OpenCV installation required)
  • Identifying specific facial features in an image
  • Digital make-up application
  • Locating and recognizing unknown faces in a picture based on previous pictures of people already known
  • Locating persons in a photo and drawing boxes around each
  • Face comparison through numeric face distance instead of just True/False matches
  • Recognizing faces in a live video from your webcam - Simple / Slow Version or Fast Version (OpenCV installation required)
  • Face recognition in a video file, and writing out a new video file (OpenCV installation required)
  • Face recognition on a Raspberry Pi w/ camera
  • Running a web service to identify faces via HTTP (Flask installation required)
  • Recognizing faces with a K-nearest neighbors classifier
  • Training multiple images per person, and then recognizing those faces through an SVM

Project Samples

Finding faces in pictures Finding and manipulating facial features in pictures Applying digital make-up Identifying faces in pictures

Project Activity

See All Activity >