Kenyan Sign Language Classification

Welcome to the Kenyan Sign Language (KSL) Classification repository! This project focuses on developing a machine learning model to recognize ten everyday KSL signs in images, addressing the bias present in many sign language datasets. The model was developed using Jupyter Notebook, and the dataset was collected by Task Mate for a Zindi competition.

Table of Contents

1. Introduction
2. Dataset
3. Model Development
4. Getting Started
5. Contribution Guidelines
6. License

Introduction

The Kenyan Sign Language Classification project aims to provide an accurate and unbiased recognition model for KSL signs. With a focus on inclusivity, this project helps address the underrepresentation of people of color in sign language datasets, ensuring a more equitable solution for KSL recognition.

Dataset

The dataset used for this project was collected specifically for a Zindi competition by Task Mate. It consists of images featuring ten different everyday KSL signs, with a focus on hands of people of color. This approach helps to address bias in sign language datasets and contributes to a more inclusive model.

Model Development

The KSL classification model was developed using Jupyter Notebook and relies on machine learning or deep learning algorithms to accurately recognize the ten KSL signs present in the dataset. The model was trained and validated using a portion of the dataset and achieves high accuracy in KSL sign recognition.

Getting Started

To get started with the Kenyan Sign Language Classification project, follow these steps:

1. Clone the repository using git clone https://github.com/Briankim254/Kenyan-signLanguage-Classification.git
2. Navigate to the project directory using cd Kenyan-signLanguage-Classification
3. Install the required dependencies, preferably in a virtual environment.
4. Launch Jupyter Notebook using jupyter notebook
5. Open the notebook file and explore the model development process.

Contribution Guidelines

We encourage contributions to improve the KSL classification model and expand its capabilities. To contribute, please follow these steps:

1. Fork the repository and create a new branch for your changes
2. Make your changes or additions to the project
3. Create a pull request and wait for a review from a team member

Please ensure that your code follows best practices for code quality and documentation.

License

The Kenyan Sign Language Classification project is licensed under the MIT License. This allows for open collaboration and sharing of the project while ensuring that contributors retain ownership of their work.