Inspiration
Saw the video of tiger woods getting a DUI, as a result we started thinking about driver safety and accident prevention
What it does
tracks driver face, looks for fatigue, alert when drowsy or tired, supply large data to fleets of other companies
How we built it
React Native, React, Expo, Python, scikit-learn, PyTorch, FastAPI
Challenges we ran into
Accuracy of our own model, using feature values of first model to train second model, coming up with our own algorithm to solve false positives
Accomplishments that we're proud of
It works
What we learned
Resources and time needed to train models, how certain models work and operate
What's next for Heimdall
Deploy on first wave of test drivers
Built With
- css
- expo.io
- fastapi
- mediapipe
- python
- pytorth
- react
- reactnative
- scikit-learn
- tailwind
- vite
Log in or sign up for Devpost to join the conversation.