3D Particle Cloud
A 3D particle cloud renders thousands of points as a single GPU-instanced draw, with each particle's xyz position drawn from the data and color encoding either a category or a continuous metric. Useful for sensor sweeps, point-cloud scans, model embeddings (e.g. UMAP/t-SNE projec
Template details
- chart
- 3d
- particles
- threejs
- point-cloud
- embedding
When to use this template
This visualization excels at revealing the overall shape, density, and emergent clusters within very large, multi-dimensional datasets. Analysts use it to explore complex relationships and identify structural patterns in data where individual point details are less critical than the collective distribution. Its interactive 3D navigation allows for inspecting cluster boundaries and understanding spatial relationships from various perspectives.
- Datasets with thousands to millions of individual records
- Each record having at least three continuous numeric dimensions for position
- Optional fourth dimension for color encoding (categorical or continuous)
- Data where overall density and cluster formation are primary insights
- Datasets representing spatial coordinates or abstract embeddings
- Data with a high signal-to-noise ratio in its spatial distribution