This is a Python code that utilizes the face_recognition library and OpenCV to perform real-time face recognition on a video feed from a webcam. The code matches the detected faces with pre-trained known faces and displays the name of the recognized person on the video frame.
Before running this code, make sure you have the following installed:
- Python 3.x
- OpenCV (cv2) library
- face_recognition library
You can install the necessary libraries using pip:
pip install opencv-python
pip install face-recognition
-
Clone the repository or download the source code file.
-
Run the following command in your terminal to execute the code:
python face_recognition.py -
The program will access the webcam and display the video feed.
-
The code loads pre-trained face encodings and names from the
known_face.picklefile. Make sure you have this file in the same directory as the script. -
As faces are detected in the video feed, the code compares them with the known face encodings and displays the recognized names on top of the faces.
-
Press 'q' to quit the program.
Note: You may need to adjust the face recognition parameters according to your requirements. The current threshold for matching faces is set to 0.4 (results = face_recognition.compare_faces(encodings, encod, 0.4)).
Feel free to modify the code to fit your specific needs and integrate it into your projects!
This code was developed by Sajad Kamali.