FaceMap: ML-based Face Recognition Web Application

Project Overview

FaceMap is an advanced face recognition web application that combines the latest techniques in machine learning, image processing, and MLOps to provide accurate and automated face detection and recognition capabilities. Built with Flask and deployed on Heroku, it showcases the practical application of AI in image and video analysis.

Features and Capabilities

• Automated Face Detection: Utilizes OpenCV to detect faces in images and videos effortlessly.
• Sophisticated Face Recognition: Implements a machine learning model using Eigenfaces and PCA for feature extraction, and Support Vector Machines (SVM) for the classification of faces.
• Image Processing and Analysis: Employs OpenCV for preprocessing images to enhance model accuracy.
• Model Training and Evaluation: Offers tools for training, testing, and evaluating the model's performance to ensure reliable face recognition.
• Hyperparameter Tuning: Uses grid search to fine-tune the model for optimal results.
• Flask Web Server: A robust backend built in Python with Flask, capable of handling RESTful API requests.
• User-Friendly Interface: Features a responsive front-end design using HTML, CSS, and Bootstrap for a seamless user experience.
• REST API Creation: Enables the development of REST APIs for easy client-side integration.
• Deployment on Heroku: The application is cloud-ready and deployed on Heroku for widespread accessibility.
• Security and Privacy: Designed with a focus on user data security and privacy best practices.

Technical Stack

• Python: Backend logic and machine learning.
• Flask: Web server and API development.
• OpenCV: Image processing.
• Heroku: Cloud deployment.
• Bootstrap: Frontend design.

Screenshots

Overview in Details

SVM Model

Test Output 01

Test Output 02

Test Output 03

Classification Report

Getting Started

To set up and run FaceMap:

1. Clone the repository to your local system.
2. Install the required dependencies with pip install -r requirements.txt.
3. Configure the necessary environment variables.
4. Start the Flask server using flask run.
5. Access the web interface through your browser to upload images or videos for face recognition.

Documentation and Resources

• Project Report: For a detailed analysis of the project development and performance metrics, check out the Project Report PDF.
• Dataset: The dataset used for training the machine learning model is hosted on Google Drive and can be accessed here.

Disclaimer

This application is intended for educational and research purposes and should be used responsibly according to privacy laws and regulations.