Skip to content

Color Maps

Built-in color maps

Different color maps are appropriate for different situations:

  • sequential maps data in to a linear range (when it doesn’t fall in to one of the more specific categories below).
  • diverging maps data in to a linear range, where the center value is “neutral” and the endpoints of the range are opposite and symmetric. Examples include a rate of change, or a residual; in both cases 0 semantially means “nothing”, and the endpoints have opposite meaning.
  • cyclic maps data defined on the circle, like an angle
  • decorative maps should generally not be used to encode numerical data, but may be useful for other visualization purposes

Polyscope supports the following built-in color maps:

Name Type String Key
viridis sequential viridis
blues sequential blues
reds sequential reds
coolwarm diverging coolwarm
pink-green diverging pink-green
phase cyclic phase
spectral decorative spectral
rainbow decorative rainbow
jet decorative jet
turbo decorative turbo

Loading custom color maps

Custom colormaps can be loaded at runtime from image files and used anywhere colormaps are used. Loading can be performed with the UI from [Appearance] --> [Color maps] --> [Load color map], or programatically using the function below. The input should be a horizontally-oriented image file like the one below; the centermost row of pixels will be read to generate the color map. Most common image formats are accepted (anyting stb_image can read).

sample colormap

load_color_map(cmap_name, filename)
load_color_map

Load a new colormap from the image file filename, which will be assigned the name cmap_name.

Example:

import polyscope as ps

ps.load_color_map("fancy cmap", "my_cmap.png")


  1. Viridis is by Nathaniel J. Smith, Stefan van der Walt, and Eric Firing. link 

  2. Phase is from the cmocean package. link 

  3. Turbo is by Anton Mikhailov and others at Google. link 

  4. The other color maps have unclear origins or are simple linear ramps, and are implemented in matplotlib