Interactive GUI toolkit for robotics visualization - Python & C++, runs on desktop and web

Developer Pascal Thomet has released Dear ImGui Bundle after four years of solo development, creating a comprehensive open-source framework for robotics visualization. The toolkit wraps the popular Dear ImGui library with 23 integrated components including ImPlot for 2D/3D plotting, ImmVision for camera feed inspection with zoom/pan capabilities, and various node editors and 3D gizmos. Built in both Python and C++, it delivers real-time visualization at 60-120fps with GPU acceleration via OpenGL/Metal/Vulkan, making it ideal for sensor data monitoring, trajectory plotting, and live parameter adjustment.

The framework's immediate mode architecture eliminates traditional GUI boilerplate, allowing developers to create interactive interfaces with just a few lines of code. It provides real-time sliders, knobs, toggles, and color pickers that update instantly without callback complexity. Crucially, the same codebase deploys across Linux, macOS, Windows, and even browsers via Pyodide (Python) or WebAssembly/Emscripten (C++), enabling easy sharing of robotics dashboards without installation requirements.

Already adopted in several research projects including CVPR 2024 papers (notably the 4K4D project), Newton Physics, and moderngl, the toolkit has gained over 1,100 GitHub stars. The Python bindings are auto-generated using Thomet's litgen tool, ensuring they stay synchronized with upstream Dear ImGui developments. The framework's robotics applications include live camera feed processing with synchronized zoom views, real-time filter parameter adjustment, and cross-platform deployment of visualization tools that can integrate with ROS workflows.