Skip to content

Latest commit

 

History

History

Background_Colorization

Folders and files

NameName
Last commit message
Last commit date

parent directory

..
data
data
 
 
data_preparation
data_preparation
 
 
data_processing
data_processing
 
 
.gitignore
.gitignore
 
 
README.md
README.md
 
 
bg_colorization_main.py
bg_colorization_main.py
 
 

README.md

Background Colorization Module

This directory hosts the code and dataset for Background Colorization Module in the SketchyScene Colorization system.

example

Requirements

  • Python 3
  • Tensorflow (>= 1.3.0)
  • PIL

Preparations

  • Please follow the instructions here for BACKGROUND dataset preparation.

Training

After the preparations, run:

python3 bg_colorization_main.py --mode 'train'

Check the logs under outputs/(time-stamp)/log for the changes of the losses.

Testing

When the training finishes, the checkpoints can be found under outputs/(time-stamp)/snapshot. We have also provided our trained model, which can be downloaded here.

Then run the following command:

python3 bg_colorization_main.py --mode 'test' --resume_from '2019-00-00-00-00-00'
  • Set the time-stamp at resume_from to the model you want.

Results can be found under outputs/(time-stamp)/results.

Citation

Please cite the corresponding paper if you found the datasets or code useful:

@article{zouSA2019sketchcolorization,
  title = {Language-based Colorization of Scene Sketches},
  author = {Zou, Changqing and Mo, Haoran and Gao, Chengying and Du, Ruofei and Fu, Hongbo},
  journal = {ACM Transactions on Graphics (Proceedings of ACM SIGGRAPH Asia 2019)},
  year = {2019},
  volume = 38,
  number = 6,
  pages = {233:1--233:16}
}