skullphone-izer

Make images into the style of Skullphone's "digital media painting" artwork.

Here is an example of the current input/output of the notebook:

Usage:

In Terminal:

• Place image and skullphonize_image.py in the same directory.
• Run the script with python (requires NumPy, OpenCV2, MatPlotLib, PIL, sys, os), with the image and scale as arguments: python skullphonize_image.py [str IMAGE_NAME] [int SCALE]
• If a readable image is given, MatPlotLib will open a window showing input and output images side-by-side. The output image will also be saved to the cwd under the same name with scale appended.
• Errors will be printed to the terminal.

In Jupyter:

• Just open and run the notebook cell-by-cell. This is the easiest way to see how the code works.

TODO:

• Write tests. How do you write tests for image processing scripts?
• Build out GUI for input/output?
• Move to webapp (Docker based?) (shelved 2/21)
• Expand to process video in real-time (ish). Use to build a "mirror".