This repository provides a collection of C++ libraries for creating data visualization and basic UI applications, primarily focused on use cases in robotics.
The core component of this codebase is a library called "imview", which is designed to be both flexible and lightweight:
- For data visualization, imview provides a set of API functions to easily plot 2D time-series data, draw 2D primitives and render 3D objects in real-time.
- For UI design, imview includes automatic layout management and widgets such as buttons, sliders, and text boxes.
The design of imview is documented in docs/imview_design.md. If you are interested in using the imview library in your own project, it's recommended to read this design document first. Additionally, an app named "quickviz" is included with commonly used data visualization functions (to support development of libxmotion). It also serves as an example to demonstrate the usage of the imview library.
The code in this repository should build on any recent linux distributions with a compiler supporting C++11/14/17. Note that the yoga library for layout management requires certain C++20 features, and you will need to disable automatic layout feature if you are using an older compiler (such as the gcc that comes with Ubuntu 20.04 or older).
Setup toolchain
$ sudo apt install build-essential cmake
If the version of cmake bundled with your system is too low, you can install a newer version from Kitware PPA or build and install from source.
Install dependencies
$ sudo apt-get install libgl1-mesa-dev libglfw3-dev libcairo2-dev
Please refer to CI configuration for the up-to-date dependency installation instructions.
Download code
$ git clone --recursive https://github.com/rxdu/quickviz.git
Or you can clone and then update the submodules manually
$ git clone https://github.com/rxdu/quickviz.git
$ cd quickviz
$ git submodule update --init --recursive
Configure CMake and compile
$ cd quickviz
$ mkdir build && cd build
$ cmake ..
$ make -j8
(Windows support is experimental and may not be as stable as the Linux setup.)
On Windows, vcpkg is used to manage dependencies. The following instructions are tested on Windows 11 with Visual Studio 2022.
Setup toolchain
Make sure both git and cmake are added to the system PATH. This should be done automatically for you if you use an installer with default installation settings. Otherwise, you can add the paths manually if you prefer installing them with binary archives.
Install vcpkg
Refer to the vcpkg documentation for more details.
$ git clone https://github.com/microsoft/vcpkg.git
$ cd vcpkg
$ .\bootstrap-vcpkg.bat # if you are using powershell
Make sure the following two environment variables are set:
export VCPKG_ROOT=/path/to/vcpkg
export PATH=$VCPKG_ROOT:$PATH
You can add the two variables in "Environment Variables" in Windows settings to make them persistent. Note that you should use
%VCPKG_ROOT% instead of $VCPKG_ROOT when appending it to the PATH variable.
Build package
Depending on the IDE you use, the configuration steps may vary.
For Visual Studio, open "Build"-"Cmake Settings for quickfiz" and add the following line to "CMake command arguments":
-DCMAKE_TOOLCHAIN_FILE=<your-vcpkg-path>/scripts/buildsystems/vcpkg.cmake
Then you should be able to configure and build the project successfully.
The library is built on top of a few third-party libraries, you can refer to their documentation for more details:
- imgui demo: src/third_party/imcore/imgui/imgui_demo.cpp
- implot demo: src/third_party/imcore/implot/implot_demo.cpp
- cairo docs: https://cairographics.org/documentation/
- yoga layout docs: https://www.yogalayout.dev/docs/styling/
Online demo of imgui and implot: