Python Face Recognition

Facia
January 20, 2025
Biometric News, Biometric Updates and News

Python Face Recognition is a deeply used method to identify or verify individuals in images or videos. Python offers several libraries and frameworks to implement Face Recognition Systems. Below is an overview of popular tools and steps to perform face recognition in Python.

Popular Libraries for Face Recognition in Python

Face_recognition

Built on dlib, a C++ library.
Easy to use and highly accurate.
Can perform face detection, Face Recognition, and even Facial Feature extraction.

Installation:

pip install face_recognition
pip install opencv-python

OpenCV

An open-source computer vision library.
Provides pre-trained models for face detection using Haar cascades or DNN-based models.

Dlib

Offers face detection and landmark extraction.
Provides 68 facial landmarks for advanced face processing.

DeepFace

Built on top of Keras/TensorFlow.
Supports multiple pre-trained models like VGG-Face, Google FaceNet, and OpenFace.

Basic Steps for Face Recognition

1. Install Necessary Libraries

Install the required dependencies:

pip install face_recognition opencv-python

2. Code for Face Recognition

Here’s a simple implementation using the face_recognition library:

import face_recognition
import cv2
Load a sample picture and learn how to recognize it
known_image = face_recognition.load_image_file(“known_person.jpg”)
known_encoding = face_recognition.face_encodings(known_image)[0]
Load an image with an unknown face
unknown_image = face_recognition.load_image_file(“unknown_person.jpg”)
Find all face encodings in the unknown image
unknown_encodings = face_recognition.face_encodings(unknown_image)
for unknown_encoding in unknown_encodings:
# Compare faces
results = face_recognition.compare_faces([known_encoding], unknown_encoding)

if results[0]:
print("Match Found!")
else:
print("No Match.")

3. Real-Time Face Recognition with OpenCV

For real-time recognition using a webcam:

import face_recognition
import cv2
Load a known face
known_image = face_recognition.load_image_file(“known_person.jpg”)
known_encoding = face_recognition.face_encodings(known_image)[0]
Initialize webcam
video_capture = cv2.VideoCapture(0)
while True:
# Capture video frame-by-frame
ret, frame = video_capture.read()

# Convert frame to RGB (OpenCV uses BGR by default) rgb_frame = frame[:, :, ::-1] # Find face encodings in the frame face_locations = face_recognition.face_locations(rgb_frame) face_encodings = face_recognition.face_encodings(rgb_frame, face_locations) for face_encoding in face_encodings: # Compare faces results = face_recognition.compare_faces([known_encoding], face_encoding) if results[0]: print("Match Found!") # Draw a rectangle around the face top, right, bottom, left = face_locations[0] cv2.rectangle(frame, (left, top), (right, bottom), (0, 255, 0), 2) # Display the resulting frame cv2.imshow('Video', frame) # Break the loop with 'q' key if cv2.waitKey(1) & 0xFF == ord('q'): break

Release the webcam and close windows
video_capture.release()
cv2.destroyAllWindows()

Advanced Features

Face Landmarks Detection
Use the face_recognition.face_landmarks() function to Detect Facial Landmarks like eyes, nose, and mouth.
Face Encoding Database
Store multiple encodes for multiple individuals and compare them with new images for better scalability.
Integration with Deep Learning Models
Combine face recognition with models like FaceNet, Dlib, or OpenCV DNN for enhanced accuracy.