Facial Landmark Detection with OpenCV

This project demonstrates how to detect facial landmarks in real-time using OpenCV. It uses a pre-trained Haarcascade classifier for face detection and a pre-trained LBF (Local Binary Features) model for facial landmark detection. The application captures frames from a webcam, detects faces and their landmarks, and displays them with connected lines and circles.

Features

• Real-time face detection using Haarcascade classifier.
• Real-time facial landmark detection using the LBF model.
• Visualization of facial landmarks with both circles (dots) and lines connecting the points.
• Saves the last captured frame with detected landmarks.

Requirements

• Python 3.7+
• OpenCV (cv2)
• NumPy
• Internet connection to download the necessary models

Setup Instructions

1. Clone the Repository:

   git clone https://github.com/yourusername/facial-landmark-detection.git
   cd facial-landmark-detection

2. Install the Required Packages:

   You can install the required Python packages using pip:

   pip install opencv-python opencv-contrib-python numpy

3. Download the Pre-trained Models:

   The script automatically downloads the Haarcascade classifier for face detection and the LBF model for facial landmark detection if they are not already present in the data directory.

Usage

1. Run the Script:

   python facial_landmark_detection.py

2. Functionality:

   • The script opens a window showing the live feed from your webcam.
   • Detected faces and their landmarks are displayed with circles and lines.
   • Press q to quit the application and close the window.

3. Saved Output:

   • The script saves the last captured frame with detected landmarks as face-detect.jpg in the current directory.

Code Overview

'''
Facial Landmark Detection in Python with OpenCV

Detection from webcam
'''

# Import Packages
import cv2
import os
import urllib.request as urlreq
import numpy as np