Deep Color Transfer using Histogram Analogy
_{Official PyTorch Implementation of the CGI 2020 Paper
Project | Paper | Supp | Slide}

This repo contains the evaluation code for the following paper:

Deep Color Transfer using Histogram Analogy
Junyong Lee¹, Hyeongseok Son¹, Gunhee Lee², Jonghyeop Lee¹, Sunghyun Cho¹, and Seungyong Lee^1
1POSTECH, ²NCSOFT
The Visual Computer (special issue on CGI 2020) 2020

Figure: Color transfer results on various source and reference image pairs. For visualization, the reference image is cropped to make a same size with other images.

Getting Started

Prerequisites

Tested environment

1. Install requirements
   • pip install -r requirements.txt

2. Pre-trained models

   • Download and unzip pretrained weights (OneDrive | Dropbox) under [CHECKPOINT_ROOT]:

     ├── [CHECKPOINT_ROOT]
     │   ├── *.pth

     NOTE:

     [CHECKPOINT_ROOT] can be specified with the option --checkpoints_dir.

Testing the network

• To test the network:

  python test.py --dataroot [test folder path] --checkpoints_dir [CHECKPOINT_ROOT]
  # e.g., python test.py --dataroot test --checkpoints_dir checkpoints

  Note:

  • Input images and their segment maps should be placed under ./test/input and ./test/seg_in, respectively.
  • Target images and their segment maps should be placed under ./test/target and ./test/seg_tar, respectively.
  • The test results will be saved under ./results/.

• To turn on semantic replacement, add --is_SR:

  python test.py --dataroot [test folder path] --checkpoints_dir [ckpt path] --is_SR

Contact

Open an issue for any inquiries. You may also have contact with junyonglee@postech.ac.kr

Resources

All material related to our paper is available via the following links:

License

This software is being made available under the terms in the LICENSE file. Any exemptions to these terms require a license from the Pohang University of Science and Technology.

Citation

If you find this code useful, please consider citing:

@Article{Lee2020CTHA,
    author  = {Junyong Lee and Hyeongseok Son and Gunhee Lee and Jonghyeop Lee and Sunghyun Cho and Seungyong Lee},
    title   = {Deep Color Transfer using Histogram Analogy},
    journal = {The Visual Computer},
    volume  = {36},
    number  = {10},
    pages   = {2129--2143},
    year    = {2020},
}