Image Inpainting Project

This project uses a simple CNN trained to predict a full image from an image that is obscured with a black grid.

Example usage

To train a new model run

python main.py working_config.json

To run the Streamlit application run

streamlit run streamlit_application.py

Structure

The folder structure of the project

Image-Inpainting-Project
|- architectures.py
|    Classes and functions for network architectures
|- datasets.py
|    Dataset classes and dataset helper functions
|- main.py
|    Main file. In this case also includes training and evaluation routines.
|- README.md
|    A readme file containing info on project, example usage and dependencies.
|- utils.py
|    Plotting functions for the training procedure.
|- working_config.json
|     Configuration file for specifying the network architecture and training parameters. Can also be done via command line arguments to main.py.
|- streamlit_application.py
|     Streamlit script. Used for project (inference) demonstration.

Dependencies

To install the project environmen, run

conda env create -f environemnt.yml