Skip to content

Latest commit

 

History

History
105 lines (67 loc) · 2.74 KB

README.md

File metadata and controls

105 lines (67 loc) · 2.74 KB
Raw

Anti-Spoofing-Liveliness-Detector

Overview

This project implements a computer vision system for Anti-Spoofing, specifically liveliness detection, using the YOLO (You Only Look Once) model. The system is designed to distinguish between real and fake faces in live webcam or images.

Table of Contents

Installation

  1. Clone the repository:
git clone https://github.com/Kaustbh/Anti-Spoofing-Liveliness-Detector
  1. Install dependencies:

    • Before Installing dependencies , create a virtual environmnet , to prevent conflicts between different projects that may have different versions of the same library.
pip install -r requirements.txt

Usage

To use the Anti-Spoofing system, follow these steps:

  1. Run the data collection script to gather real and fake face images using the webcam:
python data_collection.py
  1. Split the collected data into training, testing, and validation sets:
python split_data.py
  1. Train the YOLO model:
python train.py
  1. Run the main script to perform face detection and liveliness detection:
python main.py

Testing

The testing folder contains Python scripts and images for testing the face detection and YOLO model.

To run the tests:

python testing/facedetector.py
python testing/yolotest.py

Data Collection

The data_collection folder contains scripts for collecting real and fake face images using the webcam. Run the data_collection.py script to initiate the data collection process.

Training

The training folder contains scripts and resources for training the YOLO model. The model weights are stored in the weights folder. To train the model, use the train.py script.

python training/train.py

Folder Structure

  • data_collection/: Scripts for collecting real and fake face images using the webcam.
  • testing/: Python scripts and images for testing the face detection and YOLO scripts.
  • training/: Scripts and resources for training the YOLO model.
  • weights/: Folder where trained model weights are stored.
  • main.py: Main script for face detection and liveliness detection.
  • split_data.py: Script to split collected data into training, testing, and validation sets.
  • requirements.txt: List of dependencies.

Contributing

If you'd like to contribute to this project, please fork the repository and create a pull request. Issues and feature requests are welcome as well!

License

This project is licensed under the MIT License.