Skip to content

ColCarroll/imcmc

Repository files navigation

imcmc

Build Status Coverage Status

It probably makes art.

imcmc (im-sea-em-sea) is a small library for turning 2d images into probability distributions and then sampling from them to create images and gifs. Right now it is best at logos and shape based images.

Installation

This is actually pip installable from git!

pip install git+https://github.com/ColCarroll/imcmc

Quickstart for logos and gifs

See imcmc.ipynb for a few working examples as well.

import imcmc


image = imcmc.load_image('python.png', 'L')

# This call is random -- rerun adjusting parameters until the image looks good
trace = imcmc.sample_grayscale(image, samples=1000, tune=500, nchains=6)

# Lots of plotting options!
imcmc.plot_multitrace(trace, image, marker='o', markersize=10,
                      colors=['#0000FF', '#FFFF00'], alpha=0.9);

# Save as a gif, with the same arguments as above, plus some more
imcmc.make_gif(trace, image, dpi=40, marker='o', markersize=10,
               colors=['#0000FF', '#FFFF00'], alpha=0.9,
               filename='example.gif')

Python

Quickstart for color images

See crosshatch.ipynb for a few working examples as well.

import matplotlib.pyplot as plt

from imcmc.color import (
    ImageLines,
    IntensityMCMCStrategy,
    UniformLinesStrategy,
    GibbsIntensityStrategy
)

pete = plt.imread('color/pete2.jpg')

ImageLines(pete, UniformStrategy()).plot()

pete

munchen = plt.imread('color/munchen.jpg')

ImageLines(munchen, IntensityMCMCStrategy(step_size=500)).plot(10_000)

munchen

beach = plt.imread('color/beach.jpg')

ImageLines(beach, UniformLinesStrategy()).plot(1500, linewidth=1)

beach

karwendel = plt.imread('color/karwendel.jpg')

ImageLines(karwendel, GibbsIntensityStrategy()).plot(1_000)

karwendel

Built with

Pillow does not have a logo, but the other tools do!

PyMC3

matplotlib

scipy

Python

Here's a tricky one whose support I appreciate

I get to do lots of open source work for The Center for Civic Media at MIT. Even better, they have a super multi-modal logo that I needed to use 98 chains to sample from!

Center for Civic Media

About

Image Markov Chain Monte Carlo

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 4

  •  
  •  
  •  
  •  

Languages