A 3D point cloud is a collection of measured locations in three-dimensional space. Each point has coordinates and may carry attributes such as color, timestamp, or intensity. Point clouds preserve the shape of a real environment rather than organizing it into named building components or optimized rendering surfaces. Use one when the measured geometry matters to a digital twin’s decisions. section 2.4.2 section 2.7
LiDAR is one way to capture a point cloud. It emits laser pulses, times their reflections, and calculates 3D coordinates. A scan can contain millions or billions of points. Repeated scans can reveal physical changes; infrastructure inspections, facility retrofits, and stockpile measurements are practical uses. section 7.4.5 section 2.7
Figure 2.12 A point-cloud visualization of the center of Amsterdam. Source: http://ahn2.pointclouds.nl/. Licensed under CC0 1.0 Universal (Public Domain).
The Amsterdam visualization shows how many measured points can represent a complex urban environment without first turning it into a BIM or mesh. section 2.4.2
Measure a sand pile
A LiDAR-equipped iPad and the 3D Scanner app were used to capture a sand pile and estimate its volume at approximately 1.6 cubic meters. The same measurement principle applies to ore stockpiles at industrial scale, but the required sensor accuracy and capture equipment differ. section 7.4.5
Choose the representation for the task. Point clouds suit high-accuracy as-built documentation and structural inspection; mesh-based models favor efficient visualization, while BIM adds structured information about components. Accuracy depends on the sensor: consumer devices may deliver centimeter-level results, whereas survey-grade terrestrial scanners can reach millimeter-level precision. Large point clouds also demand higher-performance hardware, and high-precision capture equipment can be expensive. Align scans with other spatial data using appropriate coordinate systems and transformations. section 7.6 section 7.4.5 section 2.7
