AR Hiking Monocle

One-eyed AI-powered vision that enhances your hiking experience.

Comment

Inspiration

Inspired by Augmented Reality Glasses from real life and movies alike

What it does

The AR Hiking Monocle acts as a one-eye wearable display. It shows you your surroundings through a live camera feed, and overlays useful hiking information: navigation hints, weather, altitude, object recognition, night vision, and more.

How we built it

We used OpenCV to handle the live camera feed and UI overlays. MobileNet SSD AI Model to detect people, animals, and certain objects. DeepLabV3 MobileNetV3 Model to highlight large bodies of water. Custom UI layers handle zoom, compass, map display, HUD labels, and night vision processing. Recording is done through OpenCV’s VideoWriter. Hardware was 3D printed and build using an Arduino Kit

Challenges we ran into

The 3D Printed material was too light and flimsy. The camera did not offer orientation or GPS information so our information had to be made up. The distance sensors were too short.

Accomplishments that we're proud of

Night vision somewhat working, and the AI recognition working. Getting the video feed to not have slow FPS. Utilizing Servos to move camera up.

What we learned

How to integrate multiple ML models into a live video. HUD/UI Design. How to model and 3D print and Arduino knowledge. How to use OpenCV.

What's next for AR Hiking Monocle

Correct information in the future when we get a camera that sends back GPS information where we can use API calls to fetch data, altitude, and AQI.

Distance checker

Friends GPS Markers

Health Monitors

Built With

  • deeplabv3
  • mobilenetssd
  • python
  • torch
Share this project:

Updates