Colorize black and white photos.
Switch branches/tags
Nothing to show
Clone or download
Permalink
Failed to load latest commit information.
data
images
DESCRIPTION.yaml Minor user experience update. Nov 14, 2018
Makefile
README.md Credit the model author. Nov 8, 2018
configure.sh
demo.py
display.py Refactor code for better capturing model file not found error Nov 14, 2018
print.py
score.py 🐛demo and score failed at 'predict(gray)' Nov 16, 2018
utils.py

README.md

Simple Colorization

This pre-built model from Yang Liu provides a very simple example of a photo colorization using deep learning. A collection of black and white photos is provided and the demo command will colorize and display these. You can then colorize your own local photo file, a photo at a URL, or a folder of photos using the score command.

See the github repository for examples of its usage: https://github.com/mlhubber/colorize

Usage

  • To install and run the pre-built model:

    $ pip install mlhub
    $ ml install colorize
    $ ml configure colorize
    $ ml demo colorize

Dependencies

Examples

  • To colorize:

    • An image from a local file:

      $ ml score colorize ~/.mlhub/colorize/images/image_07_bw.png

      Then the colorized image will be saved into a file like image_07_bw_color.png.

    • Images in a folder:

      $ ml score colorize ~/.mlhub/colorize/images
    • An image from the web:

      $ ml score colorize https://github.com/foamliu/Simple-Colorization/raw/master/images/sample.png
    • Interatively without repeatedly reloading the model:

      $ ml score colorize