Python Face Recognition with OpenCV and face_recognition Library

This article explains how to build a simple face‑recognition application in Python. It covers the necessary environment setup, required libraries, data preparation, and the core code that captures video, encodes known faces, and identifies faces in real time.

System preparation : Install the required packages via pip.

<code>pip install cmake
pip install dlib
pip install face_recognition
pip install numpy
pip install opencv-python</code>

If installing dlib fails, download the wheel manually and install it, for example:

<code>cd C:\Users\Dhanush\Downloads\
pip install dlib</code>

Place the Haar cascade XML file ( haarcascade_frontalface_default.xml ) in the same directory as the script.

Training the system : Create a folder named Faces in the script directory and store images of each person, naming each image with the person's name. The program will use these images to learn face encodings.

Face recognition code (the main script):

<code>import face_recognition as fr
import os
import cv2
import numpy as np
from time import sleep

face_cascade = cv2.CascadeClassifier('haarcascade_frontalface_default.xml')
cap = cv2.VideoCapture(0)
_, img = cap.read()

def get_encoded_faces():
    """Walk through the ./faces folder and encode all face images.
    :return: dict of {name: encoding}
    """
    encoded = {}
    for dirpath, dnames, fnames in os.walk("./faces"):
        for f in fnames:
            if f.endswith('.jpg') or f.endswith('.png'):
                face = fr.load_image_file('faces/' + f)
                encoding = fr.face_encodings(face)[0]
                encoded[f.split('.')[0]] = encoding
    return encoded

def unknown_image_encoded(img):
    """Encode a face given the file name"""
    face = fr.load_image_file('faces/' + img)
    encoding = fr.face_encodings(face)[0]
    return encoding

def classify_face(im):
    """Find all faces in an image and label them if known.
    :param im: str of file path
    :return: list of face names
    """
    faces = get_encoded_faces()
    faces_encoded = list(faces.values())
    known_face_names = list(faces.keys())
    face_locations = face_recognition.face_locations(img)
    unknown_face_encodings = face_recognition.face_encodings(img, face_locations)
    face_names = []
    for face_encoding in unknown_face_encodings:
        matches = face_recognition.compare_faces(faces_encoded, face_encoding)
        name = "Unknown"
        face_distances = face_recognition.face_distance(faces_encoded, face_encoding)
        best_match_index = np.argmin(face_distances)
        if matches[best_match_index]:
            name = known_face_names[best_match_index]
        face_names.append(name)
    for (top, right, bottom, left), name in zip(face_locations, face_names):
        cv2.rectangle(img, (left-20, top-20), (right+20, bottom+20), (255, 0, 0), 2)
        cv2.rectangle(img, (left-20, bottom-15), (right+20, bottom+20), (255, 0, 0), cv2.FILLED)
        font = cv2.FONT_HERSHEY_DUPLEX
        cv2.putText(img, name, (left-20, bottom+15), font, 1.0, (255, 255, 255), 2)
    while True:
        cv2.imshow('IMAGE', img)
        return face_names

print(classify_face("test"))
</code>

Output verification : Running the script opens the webcam, captures frames, and displays the detected faces with bounding boxes and labels. Unknown faces are marked as "Unknown".

Conclusion : The guide provides a complete, runnable example for face detection and recognition in Python, and points readers to additional resources for real‑time video processing.