Hand-Sign Recognition Model Project

This project guides you through the process of collecting images, creating a dataset, training a hand-sign recognition model, and testing it. Follow the steps below to set up and run the project.

Prerequisites

• Python 3.x installed on your system
• Required dependencies (see requirements.txt for details)
• A webcam or image files for collecting face data

Installation

Clone this repository to your local machine:

git clone https://github.com/sahilmurhekar/signlanguage
cd signlanguage

Streamlit App

Install the requirements and test the app on Streamlit:

pip install -r requirements.txt
streamlit run app.py

Main Workflow

1. Run the following script to capture images of your face and create a dataset

python collect_images.py

2. Run the script below to preprocess the collected images and generate a data.pickle file

python create_dataset.py

3. Train the face recognition classifier using

python train_classifier.py

4. Run the inference script to test the trained model

python inference_classifier.py

Notes

1. Ensure your webcam is connected (if applicable) when running collect_images.py.
2. The alternative Streamlit method (app.py) provides a user-friendly interface and may require additional setup depending on your environment.