ScaleFree: Dynamic KDE for Multiscale Point Cloud Exploration in VR

| | |:--------------------------------------------------------------------------------------------------------------------------:|

Arxiv | Video(coming soon) | Paper(coming soon)

ScaleFree is a a GPU-accelerated adaptive Kernel Density Estimation (KDE) algorithm for scalable, interactive multiscale point cloud exploration. Refer to the Video(coming soon) for a quick review. For more details, please refer to our Paper(coming soon) (ScaleFree: Dynamic KDE for Multiscale Point Cloud Exploration in VR), which will be presented on IEEE VR 2026.
Any pull requests and issues are welcome. If you find it useful, could you please give a star? Thanks in advance.

Requirements

• Windows platform
• Unity 3D (version>2019)

Install the project and Play the demo

• Download Unity Hub. Please refer to sec.1-4 in tutorial if you are a new Unity user.
• Clone the repo or download the archive https://github.com/ScaleFree/archive/refs/heads/master.zip and open the project using Unity (versions equal to/higher than 2019 have been tested). Please refer to sec.6 in tutorial if you don't know how to open an existing project.
• Assets/PointCloud-Visualization-Tool/scenes/KernelDensityEstimation.unity is the demo scene.

Control

• To select the dataset, click the Data gameObject in hierarchy and change variable datasets in project in the inspector window.
• Start the game and you can see the real-time iso-surface reconstruction of the density estimation as the following figure (Marching Cubes).
• To change MarchingCube threshold, unfold the Data gameObject in hierarchy, click MarchingCube child gameObject and adjust the variable MC Threshold in the inspector window.

BibTex