Interactive tool for GAN Inpainting

An interactive tool with an implementation of the Generative Image Inpainting algorithm found here. This an interactive website for demonstration of inpainting results using models trained on various image data sets. This demo be used for academic purposes only.

Prerequisites:

• Bash Shell
• Anaconda

Preparing the environment:

Run the following commands in your shell:

conda config –add channels conda-forge
conda install tensorflow
conda install opencv
conda install pyyaml
pip install git+https://github.com/JiahuiYu/neuralgym
conda install pybottle

Running the server:

The backend has the following directory structure:

├── Interactive_tool-for-GAN-Inpainting
│   ├── server.py
│   └── test.py
│   └── inpaint.yml
│   └── inpaint_ops.py
│   └── inpaint_model.py
│   └── input.png
│   └── inputA.png
│   └── inputB.png
│   └── output.png
│   └── output.png
│   └── output2.png
│   └── model_logs
│	│   └── release_imagenet_256
│	│   └── release_places2_256
│	│   └── release_celeba_256
│	│   └── release_dtd_256
│   └── neuralgym_logs
│	│   └── <logs will go here>
│   └── views
│	│   └── index.html
│	│   └── style.css
│	│   └── banner.png
│   └── data
│	│   └── <static input data>

To run the server, run the following commands in the root directory:

python server.py

The webpage can now be accessed at http://localhost:8080/home/index.html