Skip to content

Use PCA analysis to reorganize the pixels of a painting into a smooth color palette.

License

Notifications You must be signed in to change notification settings

ardila/paintingReorganize

Repository files navigation

paintingReorganize

To install

pip install -r requirements.txt

If that doesn't work you will have to manually track down and install these pip packages.

To use

python palette.py /path/to/input_file.jpg

Output will be written to output.png

Examples

Input:

Kupka - Mme Kupka Among Verticals

Input

Output:

Output

Input:

Van Gogh - Starry Night

Input

Output:

Output

Input:

Picasso - Demoiselles D'Avignon

Input

Output:

Output

Input:

Cezanne - The Large Bathers

Input

Output:

Output

How it works

Mostly based on PCA a little tutorial on what that means here

  1. Organize pixels according to their first principal component (the direction along which variance is greatest)

  2. Grab a column of pixels at a time, and organize them according to their first principal component

  3. Use the resulting columns one by one in the final result. With each column, check to see if flipping it upside down gives a better match to the previous column.

About

Use PCA analysis to reorganize the pixels of a painting into a smooth color palette.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages