Image Harmonization

PyTorch implementation for image harmonization.

Prerequisites

• Pytorch
• MS COCO Dataset

Getting Started

Installation

• Install PyTorch and dependencies from http://pytorch.org
• Install Torch vision from the source.

git clone https://github.com/pytorch/vision
cd vision
python setup.py install

• Install python libraries visdom and dominate.

pip install visdom
pip install dominate

Harmonization Dataset

• To train the model, generate a harmonization dataset with MS COCO
• The harmonization dataset will be randomly generated by converting brightness, contrast, and gamma value of masked region.

cd data_gen
python data_gen.py --image_dir coco_dataset_path --save_dir directory_for_saving_images 

Train

• Train a model:

python train.py --dataroot ./datasets 

• To view training results, run python -m visdom.server and click the URL http://localhost:8097. To see more intermediate results, check out ./checkpoints/web/index.html

Test

• Test the model:

python test.py --dataroot ./datasets

The test results will be saved to a html file here: ./results/test_latest/index.html.