What a 3D Scanner Does and How It Works

A 3D scanner captures the shape and surface of a physical object and converts it into a digital file that a computer can read and manipulate. Instead of taking a flat photograph, a 3D scanner records depth information — how far away each point on the object's surface is from the scanner itself. The result is a three-dimensional digital model you can rotate, measure, edit, or print.

The scanner does this by sending out light or sound waves, measuring how long they take to bounce back, and calculating distance from that timing. Different scanner types use different methods, but they all follow the same basic principle: measure many points across the object's surface, record their positions in 3D space, and stitch those measurements together into a single model.

Key Takeaways

  • 3D scanners work by sending out light or sound and measuring how the waves bounce back to calculate the distance to each point on an object's surface.
  • Structured light scanners project patterns onto an object and read how the pattern distorts, making them fast and accurate for small to medium objects.
  • Laser scanners measure the time it takes for a laser beam to return, working well for larger objects and longer distances.
  • Photogrammetry uses overlapping photographs and software to build a 3D model without special hardware, though it requires more processing time.
  • The final output is a point cloud or mesh file that can be edited, measured, or sent to a 3D printer.

Structured Light Scanners: Projecting Patterns to Read Depth

A structured light scanner projects a known pattern — usually a grid of lines or dots — onto the object you want to scan. A camera watches how that pattern bends and stretches as it hits the curved and angled surfaces. Where the object is closer to the scanner, the pattern compresses; where it is farther away, the pattern stretches. The scanner's software calculates the distance to each point by measuring how much the pattern has warped.

This method is fast and accurate for objects the size of a small toy to a human face. Most handheld 3D scanners and the depth cameras in smartphones use structured light. The trade-off is that structured light works best indoors or in controlled light — bright sunlight can wash out the projected pattern and ruin the scan.

Structured light scanners typically capture a scan in seconds to a few minutes, depending on how detailed you want the model to be. The resulting file is usually a point cloud — millions of individual points in 3D space — which can then be converted into a solid mesh that looks like a surface.

Laser Scanners: Measuring Time of Flight

A laser scanner fires a laser beam at the object and measures how long the light takes to bounce back. Since light travels at a known speed, the scanner can calculate the distance: if the light took 10 nanoseconds to return, the object is roughly 1.5 meters away. The scanner repeats this thousands or millions of times, moving the laser across the object's surface or rotating the object in front of the laser.

Laser scanners excel at scanning large objects and outdoor scenes because the laser beam is narrow and bright enough to work in daylight. They are common in surveying, architecture, and industrial inspection. A laser scanner can capture a building facade or a landscape from hundreds of meters away.

The downside is speed and cost. Laser scanners are slower than structured light — a detailed outdoor scan can take minutes to hours — and the equipment is more expensive. The output is also a point cloud, often with millions of points that require processing to become a usable model.

Photogrammetry: Building 3D Models From Photographs

Photogrammetry does not use a special scanner at all. Instead, you take many overlapping photographs of an object from different angles — typically 20 to 100 photos depending on the object's complexity — and feed them to software. The software finds matching details in the overlapping regions of the photos, calculates the distance to each detail based on how the angle changes between shots, and builds a 3D model.

This method works with any camera, including a smartphone. It is free or low-cost if you use open-source software like Meshroom or CloudCompare. The trade-off is processing time: a photogrammetry scan of a medium object can take 30 minutes to several hours to process on a home computer, depending on photo count and resolution.

Photogrammetry also requires good lighting and clear detail on the object's surface. Shiny, reflective, or featureless surfaces (like a white plastic ball) are difficult to scan because the software cannot find matching points between photos. It works best outdoors or in bright, even light.

What Happens After the Scan: Point Clouds and Meshes

When a 3D scanner finishes, the output is usually a point cloud — a file containing millions of individual points, each with an X, Y, and Z coordinate in 3D space. A point cloud is raw data; it shows the shape of the object but looks like a scattered collection of dots if you view it directly.

To make the point cloud into something useful, software converts it into a mesh — a surface made of tiny triangles that connect the points. A mesh looks solid and can be rotated, measured, edited, or sent to a 3D printer. Common mesh file formats are STL, OBJ, and PLY.

The conversion from point cloud to mesh is not automatic and perfect. Noise in the scan (stray points from reflections or dust) can create bumps or holes in the mesh. Most scanning software includes tools to clean up the point cloud before converting it — removing outliers, filling small gaps, and smoothing rough areas.

Accuracy, Resolution, and What Affects Scan Quality

The accuracy of a 3D scan depends on the scanner type, the object's surface, and the distance from the scanner. Structured light scanners typically achieve accuracy within 0.1 to 1 millimeter for objects a few centimeters away. Laser scanners can be accurate to a few millimeters at long range. Photogrammetry accuracy depends on photo resolution and overlap; it can range from 1 millimeter to several centimeters.

Reflective surfaces like polished metal or glossy plastic are difficult to scan because the light bounces away instead of returning to the scanner. Matte surfaces and textured materials are easier. Very dark objects absorb light and may not scan well. Transparent or translucent materials like glass or clear plastic do not scan at all — the light passes through instead of bouncing back.

The resolution of a scan — how many points are captured — affects file size and detail. A high-resolution scan of a small object can produce a file hundreds of megabytes in size. A lower-resolution scan is faster and smaller but loses fine detail. Most scanners let you choose the resolution before scanning.

Common Uses for 3D Scanning

3D scanning is used in manufacturing to inspect parts for defects, in healthcare to create custom prosthetics or dental implants, and in archaeology to document artifacts without touching them. Product designers use 3D scans to reverse-engineer existing objects or to capture the shape of a clay model before refining it in CAD software. Museums scan sculptures and historical objects to create digital archives.

In entertainment, 3D scanners capture actors' faces for visual effects and video game characters. In construction and surveying, laser scanners document existing buildings and landscapes to plan renovations or create as-built records. Hobbyists use 3D scanners to scan toys, figurines, or personal objects and print copies or modifications.

Frequently Asked Questions

Can I scan shiny or reflective objects?

Shiny surfaces are difficult because light bounces away from the scanner instead of returning. Some scanners include a spray coating that temporarily dulls the surface and makes it scannable. Alternatively, you can photograph the object with photogrammetry, which is less sensitive to reflections than structured light or laser scanning.

How long does it take to scan an object?

Structured light scanners capture a scan in seconds to a few minutes. Laser scanners take longer, from minutes to hours for large scenes. Photogrammetry requires taking many photos (5 to 30 minutes) plus processing time on a computer (30 minutes to several hours). The total time depends on object size, detail level, and the method you choose.

What file format should I use for 3D printing?

STL and OBJ are the most common formats for 3D printing. Most scanning software can export to either format. Before printing, you may need to clean up the mesh in software like Meshmixer or Fusion 360 to remove holes, thin walls, or other issues that would cause printing problems.

Do I need expensive equipment to scan objects?

No. Photogrammetry requires only a camera and free software. Some smartphones have depth sensors that can perform structured light scanning. If you want a dedicated scanner, handheld structured light scanners range from a few hundred to a few thousand dollars. Professional laser scanners are more expensive but are not necessary for most hobbyist or small business uses.

What is the difference between a point cloud and a mesh?

A point cloud is raw scan data — millions of individual points in 3D space. A mesh is a surface made of triangles that connects those points. A mesh looks solid and is ready for editing, measurement, or 3D printing. Most scanning software converts the point cloud to a mesh automatically, though you may need to clean it up first.