Inspiration
Nowadays digital gadgets are the essential parts of our everyday life. Working in front of the computer, contacting love ones via instant messaging on our smartphones become more frequent in the past decade, as well as during the current pandemic time. We notice that we may easily develop bad postures when using these gadgets, causing neck pain or bad pain. These harmful habits must be corrected as soon as possible.
What it does
A camera will be used for detecting users' posture. If bad postures are detected, users will be alerted.
How we built it
We trained a machine learning model on teachablemachine that can identify good/bad postures. If bad postures are detected, users will be alerted. A camera will be installed on the side to monitor the sitting postures of the user. The camera is connected to a computer that is running the program. If the model detects bad postures for more than 5 seconds, a notification message will be shown on the website.
Challenges we ran into
We could not find suitable data sets for training our machine learning model. We finally decided to take photos of our postures as the training data. But on the other side, the data used for training may not be comprehensive enough.
Accomplishments that we're proud of
We were able to find a useful web tool to train a machine learning model. We were also able to come up with a model that can successfully detect our postures and determine whether they are good or bad.
What we learned
We learned how amazing machine learning is capable of. Besides, we learned more about building a machine learning model as this was the first time for us to do so.
What's next for Posture detector
We can acquire more data for training by taking photos of the postures of different people in different places. We can also make a companion mobile app to remind users instead of showing an error message on a website.
Built With
- html
- javascript
- teachablemachine
Log in or sign up for Devpost to join the conversation.