Sample Jupyter Notebook that can be used together with the Amazon Sagemaker Model for detection and blurring (anonymization) of human faces and license plates in images.
Example provided by NavInfo Europe AWS Team.
Take a look on how to get started with Jupyter on AWS.
To run this Notebook you will need to use at least the configurations described below:
- Notebook instance type: ml.t2.medium
- Plaftorm identifier: Amazon Linux 2, Jupyter Lab 4 (notebook-al2-v3)
- Volume size: 5GB (EBS)
- Git repository: https://github.com/navinfoeurope/anonymizer (or your own repo)
You can also run this notebook locally using the Dockerfile from this repo.
Build and run the image.
$ docker build -t anonymizer .
$ docker run -it -v $(pwd):/app -p 8888:8888 anonymizerPaste the URL from the container Jupyter server app to your browser:
http://127.0.0.1:8888/tree?token=...