Definition

Point Cloud

A set of data points in three-dimensional space representing the surface geometry of an object or environment, typically captured by laser scanning or photogrammetry, used as the basis for as-built models and dimensional analysis.

A point cloud is a collection of discrete data points, each with (x, y, z) coordinates, that together represent the surface geometry of a scanned object or environment. It is the primary output of terrestrial laser scanning and a key output of drone photogrammetry. A typical room-scale laser scan might produce 50-200 million points; a large building survey might accumulate several billion points across multiple scan setups.

Each point in the cloud represents a measured location on a surface — a wall, floor, column, or beam face — as determined by the scanner or photogrammetric reconstruction. The density of points (points per square metre) determines how fine the surface detail is captured. High-density scans can resolve features as small as 1-2mm; lower-density scans used for large-scale building surveys are typically accurate to 10-20mm.

For building surveyors and structural engineers, point clouds are most commonly used to produce as-built floor plans and sections — the scanner captures the existing geometry precisely, and the point cloud is used to draw accurate plans rather than relying on tape measurements. This is particularly valuable for complex or curved geometries where manual measurement would be impractical, and for large buildings where the efficiency gain over manual methods is significant.

Point clouds are also used for deformation monitoring — comparing scans taken at different points in time to detect movement, settlement, or dimensional change. This requires careful registration of successive scans to a common coordinate reference, and careful attention to sources of error, but can detect movements as small as 5-10mm in controlled conditions.

The relationship between point clouds and photographic condition surveys is complementary. A point cloud gives you precise three-dimensional geometry but limited visual information — the colour and texture of surfaces is approximated, and fine surface features like cracks are often below the resolution threshold of the scan. A 360° photographic survey gives you rich visual information but approximate spatial position. Combining both — using the point cloud for spatial reference and the 360° photos for visual condition assessment — produces the most complete condition record available with current technology.

Related Terms

Related Pages

Put This Into Practice

pin360 lets you pin 360° photos directly onto PDF floor plans — making every survey spatially navigable. Used by structural engineers and building surveyors.

Start free