Posture Checker

A simple webcam-based posture detection app

---

Overview

This project uses pose landmark detection to analyze the user's posture and provide real-time feedback. It is designed to be run in the background to help the user maintain good posture while working for long periods at their computer.

---

Setup

Requirements

• Python 3.10 (required)
• A webcam

Install Python

Option 1: Download From python.org

Download Python 3.10 from the official website:
https://www.python.org/downloads/latest/python3.10/

Option 2: Install With Homebrew (macOS)

brew install python@3.10

Verify Python Installation

which python3.10 
python3.10 --version 

Set Up a Virtual Environment

Create the Virtual Environment:

python3.10 -m venv venv

Activate It:

On macOS / Linux:

source venv/bin/activate 

On Windows:

venv\Scripts\activate 

Install Dependencies

python -m pip install --upgrade pip
python -m pip install -r requirements.txt

---

How to Use

Step 1: Start the Program

python src/main.py

Upon startup, the setup menu will appear.

Step 2: Position Yourself

Ensure:

• Your shoulders, neck, and head are all fully visible
• You are square to the webcam

You should see pose landmarks overlaid on the webcam feed.

Step 3: Set Default Position

Sit in a comfortable, upright position and click "Set"

Your default posture is now set, and the program will begin detecting. If you intentionally slouch the program should detect your poor posture.

Step 4: Adjust Sensitivity

Use the slider to adjust the sensitivity value. This will differ from user to user. Find a comfortable value that suits you.

You may repeat Step 3 at any time to set a new default posture.

Step 5: Start Program

Press "Start"

The program will now begin running and the setup window will minimize. If the program detects poor posture while you are working, a pop-up warning message will appear in the top left coner of your screen. This message disappears when posture is corrected.

Step 6: End Program

Restore the window and click "End" to close the program.

Notes

For the best performance, it is recommended to keep the window minimized until you wish to close the program.

Lighting and camera placement can affect detection accuracy.

---

Acknowledgments

This project uses MediaPipe, licenced under the Apache License 2.0
https://github.com/google/mediapipe

---