Introduction

cwipc is a cross-platform suite for working with dynamic point clouds. It provides:

  • A language‑neutral core representation (cwipc objects) with bindings for C, C++, Python and C#.

  • Capturers for Azure Kinect, Intel RealSense, and Orbbec Femto depth cameras, supporting multi‑camera setups with registration and synchronization.

  • Compression codecs (cwipc_codec) for real‑time transmission.

  • A collection of command‑line utilities (cwipc view, cwipc grab, cwipc copy, cwipc forward, etc.) for capture, playback, streaming and analysis.

Use cases

  1. Real‑time capture and streaming – build applications that acquire point clouds from hardware, compress them and send them over the network (for example live telepresence or social XR).

  2. Offline processing – record camera streams, replay them, convert between formats, and apply filters/transformations programmatically.

  3. Research and analysis – use Python/Jupyter notebooks with NumPy, PyOpen3D or other libraries to study 3D data, machine learning models, or dataset creation.

  4. Unity integration – develop immersive experiences using the cwipc_unity package to ingest and render point clouds from C#.

  5. Cross‑language pipelines – the opaque core object can be passed between components written in different languages with minimal copying.

The library is suitable for prototyping, experimentation, and production software on Windows, macOS, Linux and Android devices.