Standing Tall

Inspiration

Using a laptop and sitting down at a desk has become an ingrained part of most people's lives, especially students. However, checking on your posture is not. With the numerous complications and health effects that bad posture can lead to, we wanted to create an elegant system to provide feedback for users' postures at the comfort of their own desks.

What it does

StandingTall includes a status bar to provide feedback on the users' postures and alert them whenever bad posture is detected.

How we built it

The user will turn on their webcam while they are working at their desk. We used OpenCV and MediaPipe to analyze the coordinates/measurements of the users' shoulders that sends an alert whenever the measurements fall out of a certain threshold.

Challenges we ran into

• Implementing the webcam and OpenCV in a flask web framework
• How we can accurately detect bad posture, even when the user sits closer to the camera or changes position
• Integrating Javascript in a Flask environment

Accomplishments that we're proud of

We were able to detect when a user's posture is bad and implement OpenCV and live streaming using a webcam through Flask.

What we learned

This was everyone's first time using Flask, as well as OpenCV and MediaPipe. We also learned just how advanced and interesting machine learning is!

What's next for Standing Tall

We hope to provide a more accurate way to provide feedback for users' postures, either through machine learning or by including more complex measurements (e.g: angles between the chin and shoulder) to detect posture.

Built With

• css
• flask
• html
• javascript
• mediapipe
• opencv