Handy Mandy

Webcam Gesture Recognition Model Instructions

This document provides instructions for setting up and running the gesture recognition model on your webcam

Prerequisites

• Python 3.7 or higher (Python 3.11.9 was used)
  • Download Python

Instructions for running the model

(Optional) It is recommended to install all Python dependencies in a virtual environment¹, but you can skip this step

Windows Powershell:

python -m venv myenv
myenv\Scripts\activate
cd "myenv"

MacOS/Linux bash:

python3 -m venv myenv
source myenv/bin/activate
cd "myenv"

1. Unzip the folder into virtual environment

2. Install all dependencies

Windows Powershell:

cd "Handy Mandy"
pip install -r .\requirements.txt

MacOS/Linux bash:

cd "Handy Mandy"
pip install -r ./requirements.txt

3. Adjust the settings on the model

Webcam dimensions(Line 7, webcam.py)

wCam, hCam = 1280, 720

4. Run the model (Note: this will not ask for permission before running your webcam!)

Windows Powershell:

python3 .\webcam.py

MacOS/Linux bash:

python3 ./webcam.py

5. If left/right hand is not accurately determined, check the following settings in the model:

Comment out Line 24, webcam.py

img = cv2.flip(img, 1)

Footnotes

1. Best practice is to use virtual environments to manage dependency conflicts between projects ↩