Posture Detection and Correction

Website which is used to correct the posture of the user sitting in front of their screen even when the process is running in the background with voice alerts

• The deployed website can be seen here

Table Of Contents:

Installation

Usage

Working

ToDo

Contributing

Installation

1. Download and install Python3 from here

2. I recommend using virtualenv. Download virtualenv by opening a terminal and typing:

   pip install virtualenv

3. Create a virtual environment with the name venv.

   • Windows

   virtualenv venv
   cd venv/Scripts
   activate

   • Linux:

   source venv/bin/activate

4. Clone this repository, extract it if you downloaded a .zip or .tar file and cd into the cloned repository.

   • For Example:

   cd A:\Posture_Detection_AND_Correction-main

5. Install the required packages by typing:

   pip install -r requirements.txt

Usage

• You can change the background sound of the application by just storing the sound file in the mentioned directory

se_hackethon_project/static/src/js/audofile.type

• Type the below command to run the Application. It will start the server on localhost:5000

  python app.py

• The Homepage of our Application.

  

• Providing the user with google authentication which directly creates the profile of the user in the backend. Also we have provided the user with traditional system of logging in registering :)

  

• After logging in the user gets to see the different options available as mentioned in the above screen shot. Lokking at some of the features.

  

  

• We don't want our users to just use our Website just as a tool, but we also want them to motivate others so that they could also make it a part of their life. So we have also created the feature of blogs in our website with basic functionality of updating deleting and so on.

  

  

  

  

• Finally when we click on Try our Website we will come to the main feature of our website :).

  

  

  

  

• Now when we click upon Analyze Session we have provided user with an interactive dashboard whenthey can see which are the most common errors they are commiting and can also print a report.

  

Working

Image Preprocessing

• So first of all we have stored all the templates in a folder

cd se_hackethon_project/templates

• After that we have created the routes of the templates in the file routes.py

• For the databse we have used sqlite and our models are stored in models.py file

• All our forms are stored the forms.py file as per the convention of flask

• All the static content need to be stored in the static folder as per the convention followed

Recognition

jeelizAr

Read about jeelizAR here

• It is basically used to recognize the face of the user
• It is used to draw the blue box around the user as seen in the above screen shots
• Also used for understanding the key points used to detect the posture

Three.js

Read about Three.js here

• As we all know that our camera does not have a depth sensing to draw a 3-D model.
• So for actually creating a 3-D model from our live-stream of web cam we use Three.js

ToDo

• Fixing the front-end of the application
• Resolve Bugs/Issues if any found.
• Optimize Code to make it faster.

Contributing

Contributions are welcome 😄

Pull requests

Just a few guidelines:

• Write clean code with appropriate comments and add suitable error handling.
• Test the application and make sure no bugs/ issues come up.
• Open a pull request and I will be happy to acknowledge your contribution after some checking from my side.

Issues

If you find any bugs/issues, raise an issue.