Home

Awesome

tinycolormap

macOS Ubuntu GitHub

A header-only, single-file library for colormaps written in C++11.

Available Colormaps

Matlab

NameSample
Parula
Heat
Hot
Jet
Gray
HSV

Reference: https://www.mathworks.com/help/matlab/ref/colormap.html

The HSV colormap is cyclic and is particularly useful for plotting angles or phases because the color transition from 1 to 0 is smooth. A phase wrap would thus not appear as a sharp edge.

Matplotlib

NameSample
Magma
Inferno
Plasma
Viridis
Cividis

These colormaps are designed to be perceptually uniform (even in black-and-white printing) and friendly to colorblindness. Cividis is specially designed such that it enables as identical interpretation to both those without a CVD and those with red-green colorblindness as possible.

Magma, Inferno, Plasma, Viridis are released under CC0 by Nathaniel J. Smith, Stefan van der Walt, and (in the case of Viridis) Eric Firing: https://github.com/BIDS/colormap/blob/master/colormaps.py. Their python code is adapted for the use in C++11.

Cividis is released under CC0 by the authors of PLOS ONE paper (Jamie R. Nuñez, Christopher R. Anderton, Ryan S. Renslow): https://doi.org/10.1371/journal.pone.0199239. We incorporated the LUT into C++11.

GitHub

NameSample
Github

This colormap is designed to mimic the color scheme used in GitHub contributions visualization.

Other

NameSample
Turbo

Turbo is developed as an alternative to the Jet colormap by Anton Mikhailov (Google LLC). See the blog post for the details. The original lookup table is released under the Apache 2.0 license. We merged it and re-licensed the part under the MIT license for consistency.

NameSample
Cubehelix

Cubehelix is developed by Dr. Dave Green and is designed for astronomical intensity images. It shows a continuous increase in perceived intensity when shown in color or greyscale. This implementation uses Green's "default" scheme (start: 0.5, rotations: -1.5, hue: 1.0, gamma: 1.0). See the original publication for details.

Dependency

tinycolormap does not have any dependencies except for C++11 standard library.

Installation

tinycolormap is a header-only library, so you do not need to compile it. You can use it by

If your project is managed by Cmake https://cmake.org/, add_subdirectory or ExternalProject_Add commands are useful as tinycolormap provides CMakeLists.txt for this purpose.

Usage

The core function of this library is

inline Color GetColor(double x, ColormapType type);

where x should be between 0.0 and 1.0 (otherwise, it will be cropped), and type is the target colormap type like Viridis (default) and Heat.

Here is a working code:

#include <iostream>
#include <tinycolormap.hpp>

int main()
{
  // Define a target value. This value should be in [0, 1]; otherwise, it will be cropped to 0 or 1.
  const double value = 0.5;

  // Get the mapped color. Here, Viridis is specified as the colormap.
  const tinycolormap::Color color = tinycolormap::GetColor(value, tinycolormap::ColormapType::Viridis);

  // Print the RGB values. Each value is in [0, 1].
  std::cout << "r = " << color.r() << ", g = " << color.g() << ", b = " << color.b() << std::endl;

  return 0;
}

Quantized colors

tinycolormap is also capable of producing quantized colormaps (i.e. the ones that have visible boundaries between colors) based on the user specified number of levels. Below is the example of the quantized Parula colormap using 10 quantization levels:

NameSample
Parula

Note that the supported range for number of levels is [1, 255].

Here is an example code that uses colormap quantization:

int main()
{
  // Define a target value. This value should be in [0, 1]; otherwise, it will be cropped to 0 or 1.
  const double value = 0.5;

  // Define number of levels for the colormap quantization. This value should be in [1, 255]; otherwise, it will be cropped to 1 or 255.
  const unsigned int num_levels = 10;

  // Get the mapped color. Here, Parula is specified as the colormap.
  const tinycolormap::Color color = tinycolormap::GetQuantizedColor(value, num_levels, tinycolormap::ColormapType::Parula);

  // Print the RGB values. Each value is in [0, 1].
  std::cout << "r = " << color.r() << ", g = " << color.g() << ", b = " << color.b() << std::endl;

  return 0;
}

Options for External Libraries Integration

Qt5 Support

When TINYCOLORMAP_WITH_QT5 is defined before including tinycolormap.hpp, for example,

#define TINYCOLORMAP_WITH_QT5
#include <tinycolormap.hpp>

(or TINYCOLORMAP_WITH_QT5 CMake option is ON), this library offers an additional utility function:

const QColor color = tinycolormap::GetColor(x).ConvertToQColor();

Eigen Support

When TINYCOLORMAP_WITH_EIGEN is defined before including tinycolormap.hpp, for example,

#define TINYCOLORMAP_WITH_EIGEN
#include <tinycolormap.hpp>

(or TINYCOLORMAP_WITH_EIGEN CMake option is ON), this library offers an additional utility function:

const Eigen::Vector3d color = tinycolormap::GetColor(x).ConvertToEigen();

Qt5 and Eigen Support

When both Qt5 and Eigen are available, this library offers additional utility functions:

inline QImage CreateMatrixVisualization(const Eigen::MatrixXd& matrix);
inline void ExportMatrixVisualization(const Eigen::MatrixXd& matrix, const std::string& path);

GLM Support

When TINYCOLORMAP_WITH_GLM is defined before including tinycolormap.hpp, for example,

#define TINYCOLORMAP_WITH_GLM
#include <tinycolormap.hpp>

(or TINYCOLORMAP_WITH_GLM CMake option is ON), this library offers an additional utility function:

const glm::vec3 color = tinycolormap::GetColor(x).ConvertToGLM();

Tools (Optional)

This repository includes the following optional tools:

Tools Build Instruction

The optional tools are managed by CMake https://cmake.org/. They can be built by, for example,

mkdir build
cd build
cmake [PATH_TO_TINYCOLORMAP] -DTINYCOLORMAP_BUILD_TOOLS=ON
make

Projects using tinycolormap

License

The MIT License (except for tools/png-exporter/stb_image_write.h, which is released under public domain).

Contribute

Pull requests are welcome.