Face Recognition for Biometric Validation

Project Overview

This project focuses on implementing face recognition for biometric validation. The scope includes understanding computational challenges in face recognition, conducting a literature review, and constructing a solution using existing methods.

Objectives

• Collect a set of face images from publicly available datasets.
• Utilize suitable methods for face recognition, including older methods or more recent deep neural network-based frameworks.
• Evaluate recognition performance using "unknown" query images and validate results with images not in the dataset.

Dataset

• The dataset should consist of more than 20 subjects/classes, each with more than 2 images.
• Images can be sourced from publicly available datasets such as LFW, CelebA, or custom datasets.

Methodology

• Implement face recognition using appropriate methods and frameworks.
• Pre-trained models on specific datasets can be utilized for deep learning-based approaches.

Tools Used

• Programming Language: Python
• Libraries: OpenCV, TensorFlow, Keras

Repository Structure

• data: Contains collected face image datasets.
• src: Source code for face recognition implementation.
• results: Stores validation results.
• README.md: Description of the project, methodology, and instructions.

Usage

1. Clone the repository.
2. Place the collected face image datasets in the "data" folder.
3. Run the source code for face recognition implementation.
4. Review the validation results in the "results" folder.

Contributors

• Shubham Vats Under the Guidance of Dr Miheryan Tuceryan( Dean Of School Of Science, Purdue University)