Inspiration
People suffer from bad posture from using electronic devices. We wanted to provide a solution that will help people fix their sitting posture using computer vision and machine learning.
What it does
Our trained model will assess your sitting position while you are on your computer. When you slouch or sit crooked for more than 5 seconds, our program will give a gentle reminder asking you to sit upright. Alongside the video, you can also view your progress graph.
How we built it
We used the teachable machine to train our model. The teachable machine is backed by Tensorflow and allows users to create and train models easily. We took this trained model and put it on a web app.
Challenges we ran into
The biggest challenge was time. We wanted to implement a native app for cross-platform functionality and a database and a platform for sharing your progress.
Accomplishments that we're proud of
The successful trained model: we manually fed different positions to truly tailor the experience for multiple sitting scenarios. Each node ensures different image degradation doesn’t affect the outcome of the model such as color, light, or the background.
What we learned
We learned about machine learning, computer vision, and teachable machine.
What's next for The Golden Ratio
At the moment, we are planning to further develop this after this hackathon and release the application on multiple platforms.
Built With
- computer-vision
- css
- html
- javascript
- machine-learning
Log in or sign up for Devpost to join the conversation.