University of AlbertaMultimedia Research Centre · Dept. of Computing Science
ROSSRemote Observation, Sensing & System
Research / R/01 Earth Observation / LiDAR and 3D reconstruction: mapping the world in points
R/01 · Earth Observation

LiDAR and 3D reconstruction: mapping the world in points

Laser scanning turns a scene into millions of 3D points. From airborne to mobile survey, the group reconstructs streets and structures for urban planning and engineering, with McElhanney.

Earth ObservationLiDAR3D ReconstructionPoint Cloud

LiDAR (light detection and ranging) measures distance by timing laser pulses as they bounce back from a surface. Sweep the beam across a scene and the returns stack into a point cloud: millions of (x, y, z) coordinates that reconstruct the world in three dimensions, without anyone setting foot on the site.

The group works across the three ways to carry the sensor:

  • Airborne laser scanning (ALS) flies the sensor on a plane or drone for wide-area, top-down coverage of terrain and forest canopy.
  • Terrestrial laser scanning (TLS) fixes the scanner on the ground to capture precise (x, y, z) coordinates of a structure or streetscape.
  • Mobile laser scanning (MLS) mounts it on a moving vehicle to survey corridors, roads and rail at speed.

Fused together, these give a 3D record that optical imagery alone cannot: depth, structure and geometry, captured day or night, independent of lighting.

Urban planning and engineering

With McElhanney in Vancouver, the group applies 3D reconstruction to urban planning and engineering. A city block is captured as a dense point cloud, then paired with optical street-view imagery of the same scene, so planners can measure geometry from the points and read context from the photograph.

A Vancouver city block reconstructed as a dense 3D LiDAR point cloud

A city block reconstructed as a dense 3D point cloud. Courtesy McElhanney.

The same Vancouver block in optical street view

The same scene in optical street view. The point cloud gives measurable geometry; the photograph gives human-readable context. Courtesy McElhanney.

Point clouds are the connective tissue of remote observation on the ground. The same representation feeds autonomous navigation, infrastructure inspection and digital-twin models, linking the aerial view from orbit to decisions made at street level.