Real-Time Facial Recognition using AI/ML

📌 Overview

This project demonstrates a real-time facial recognition system using AI/ML. It captures live video, detects faces, and recognizes identities using a TensorFlow-based model built on the VGG16 architecture.

🎯 Features

• Real-time image capture using OpenCV.
• Face detection and recognition via deep learning.
• Model optimized for fast inference with GPU support.
• Modular design for training and testing.

🛠️ Tech Stack

• Programming Language: Python
• Frameworks/Libraries: TensorFlow, OpenCV, NumPy, Matplotlib
• Model Architecture: VGG16

🚀 Installation and Usage

Prerequisites

• Python 3.8 or later
• Required libraries include TensorFlow, OpenCV, and Matplotlib. Steps to Run

1. Clone the repository and navigate to the project directory.
2. Capture images, train the model, and perform real-time recognition.

📂 Project Structure

The project includes directories for data storage, scripts for data collection and model training, and saved models for recognition tasks.

📖 How It Works

1. Data Collection: Captures images via webcam and saves them for training.
2. Model Training: Trains a facial recognition model using VGG16 for feature extraction.
3. Real-Time Recognition: Identifies faces from the live webcam feed and matches them with known identities.

📚 Future Improvements

• Add support for larger datasets.
• Implement more advanced face matching algorithms (e.g., FaceNet).
• Enhance accuracy for diverse lighting and angles.

💡 Credits

Developed by Manya Gautam as part of a real-time AI/ML project.