Map Image to Palette (vuo.image.color.palette)

Alters an image’s colors to use only colors from a given palette.

• Palette — A list of colors to allow in the output image. You can either provide a fixed list of colors, or dynamically generate it using Make Palette from Image. For colors where there’s no exact match in the palette, this node tries to find the most similar color (the color with the lowest RGB geometric distance).
• Dithering — When reducing a many-colored image to fewer colors, you may see “banding” artifacts where formerly-gradual color changes become solid regions. This port controls how to compensate for that:
  • White Noise — randomly brightens or darkens colors in the input image, producing a rough scattering effect.
  • Gradient Noise — uses a periodic pattern to brighten or darken colors, producing a smoother scattering effect. The number (0.05, 0.02, 0.01) represents the width of one noise period, in Vuo Coordinates.
  • Bayer — uses a pattern to brighten or darken colors, producing a thatched or crosshatched effect. The number represents the pattern’s tile size (e.g. for 2x2, the pattern repeats every 2 pixels on the X and Y axes). These patterns are designed to produce uniform textures, minimizing periodic artifacts, as described in Bryce Bayer’s 1973 paper, “An Optimum Method for Two-Level Rendition of Continuous-Tone Pictures”.
• Dithering Intensity — How much of the above dithering to apply to the image.

Thanks to Gisle Martens Meyer for funding development of this node!

Keywords: 1980s, 1990s, Bayer 1973, alias, apply, banding, contour, convert, crosshatch, depth, dot pattern, edge, filter, gif, gradient, gradient noise, indexed color, indirect color, match, nearest, ordered, periodic noise, perlin noise, pseudocolor, quantize, random, raster, recolor, reduce, replace, retro, scatter, thatch, white noise

Example compositions:

• Compare Posterize and Palette: Shows how the Posterize Image node's predefined, static palette compares to the Make Palette from Image node's custom palette. Image by Ashim D’Silva on Unsplash.
• Make Cyberspace Avatar: Uses live video to deep dive into Cyberspace. Thanks to Gisle Martens Meyer for funding development of some of these nodes!