Question 1

What is the story behind the median-cut color quantization algorithm?

Accepted Answer

The median-cut algorithm was introduced by Paul Heckbert in his 1982 SIGGRAPH paper as a method for choosing a limited color palette to represent an image. It works by recursively splitting the color space at the median of the channel with the widest range, producing perceptually balanced clusters. The technique became foundational in early computer graphics and is still used today for palette extraction and GIF encoding.

Question 2

How does the palette extractor decide which colors are dominant?

Accepted Answer

The tool downsamples your image to roughly 100 by 100 pixels for speed, then groups every pixel into clusters using median-cut quantization. Each cluster is split along the color channel (red, green, or blue) with the greatest range, and the average color of each final cluster becomes one swatch in your palette. The result highlights the colors that cover the most area in the image rather than the brightest or most saturated ones.

Question 3

Why does the extracted palette sometimes differ from what I expect?

Accepted Answer

Median-cut groups pixels by their numeric RGB values, which does not always match human perception of prominence. A large background area may dominate even if a smaller accent color feels more important visually. Adjusting the color count slider can help surface subtle accent tones by splitting the color space into finer buckets. Fully transparent pixels are excluded automatically so they do not skew the result.

Color Palette Extractor

About this tool

Frequently Asked Questions

Related Tools