This repository is the source code of a web app that demonstrates modern image segmentation using a 'deep learning' model in TensorFlow. The app uses the segmented output from the model to add a background blurring effect. When the input image is a portrait the application of this is to enhance the foreground-background seperation in the portrait.

The TensorFlow model is not part of this repository and the code requires an external model to be provided.

As Proof-Of-Concept the web app is deployed on the web for everyone to play with: blurbackground.online

The current deployed App utilizes the DeepLabV3 semantic image segmentation model by Google Research.

• Javascript for Web app
• React/Redux for Front End
• Razzle for Server Side Rendering
• Express for API backend
• Python for image processing
• Docker for containerization