Python based examples for the Raspberry PI AI Camera based on the IMX500 AI image sensor.
These examples are intended to be used in a virtual environment. The examples are based on the Picamera2 software.
To list the available examples:
$ ls examples/
Create a venv (virtual environment) with:
$ python3 -m venv venv --system-site-packages
$ source venv/bin/activate
$ cd examples/ # Press tab to get list of avaliable examples.
$ pip install -e .
After this step it is recommended to read the specific readme file from the examples folder if there are more steps needed or how to start the demo.
Most models in this repository are generic from the provided model zoo.
Models used in the applications are:
- Highvis - Uses a Custom trained Nanodet Model
- Line Monitoring - Uses a Custom Classification Model
- Parking Monitoring - Uses Nanodet with postprocessing included
- Queue Monitoring - Uses Nanodet with postprocessing included
- Workout Monitoring - Uses HigherhrNet, pose estimation
To convert you own models follow our MCT Tutorials to quantize your model.
Then once the model has been quantized it is time to convert it and package it to be able to run on this platform. The tutorial explains the process to convert a model_name.keras or model_name.onnx to model_name.rpk.
Please read the Site Policy of GitHub and understand the usage conditions.
