Creating a point cloud means capturing the geometry of a real place and turning it into a set of measured 3D points. There are four main capture methods and one common workflow that follows. This guide walks through both, so you can create a usable point cloud from the field to a shareable file.
Four ways to capture a point cloud
The method you choose shapes the precision, the cost, and the time on site. Here are the four that cover almost every project.
Laser scanning (LiDAR)
A terrestrial laser scanner sends laser pulses and measures the return to compute millions of precise points. It is the most accurate method for buildings and industrial sites. See our LiDAR guide and how to choose a 3D scanner.
Photogrammetry
Photogrammetry builds a point cloud from many overlapping photos, taken by camera or drone. It is affordable and great for facades and terrain, if lighting is good. See our photogrammetry software guide.
Mobile mapping and SLAM
A handheld or backpack scanner captures while you walk, using SLAM to align data on the fly. It is the fastest way to cover large or complex spaces, with slightly lower precision than a static scanner.
iPhone Pro with a mobile scanning app
An iPhone Pro equipped with LiDAR can create a point cloud quickly with a mobile app such as Mavo 3D. This method suits quick surveys and small spaces, but it is less accurate than a professional laser scanner.
| Method | Precision | Speed | Best for |
|---|---|---|---|
| Laser scanning | Very high | Medium | Buildings, industrial sites |
| Photogrammetry | Medium to high | Medium | Facades, terrain, drone |
| Mobile / SLAM | Medium | Very fast | Large or complex spaces |
| iPhone Pro / LiDAR | Low to medium | Very fast | Small spaces, quick surveys |
- A capture device: a laser scanner, a camera or drone, a mobile scanner, or a LiDAR-equipped iPhone Pro.
- Reference targets to help align scans accurately.
- Storage: raw scans are heavy, plan for enough space.
- Processing software to register and clean the data.
- A sharing platform to deliver the result to clients.
Good to know
You can combine methods on one project: a static scanner for precise rooms, a mobile scanner for corridors, and drone photogrammetry for the roof. Registration merges them into one cloud.
The point cloud workflow, step by step
- Capture: scan or photograph the site, with enough overlap between stations.
- Register: align the individual scans into a single coordinate system.
- Clean: remove noise, moving objects, and stray points.
- Export: save to a standard format such as E57 or LAS / LAZ.
- Share: upload to a cloud platform to view and share by link.
The last step, made simple
Upload your finished E57, LAS, LAZ, RCS, RCP, or LGSx file to ATIS.cloud and share it by link, up to 1 TB per file.
Which method for your project
- Interior of a building: static laser scanner for precision, mobile scanner for speed.
- Facade or heritage detail: photogrammetry, or laser for high accuracy.
- Large site or terrain: drone photogrammetry, or mobile mapping.
- Industrial plant: static laser scanning for dense, precise data.
- Tight budget: photogrammetry with a good camera.
- Quick survey of a small space: a LiDAR-equipped iPhone Pro with a mobile scanning app.
Common mistakes to avoid
- Too little overlap between stations, which breaks registration.
- Moving objects (people, vehicles) left in the data as noise.
- Poor lighting for photogrammetry, which ruins point quality.
- No reference targets, making alignment harder and less accurate.
- Skipping cleanup, which passes noise on to every later step.
Use and share your point cloud
Once the cloud is created and exported, the value comes from using it. Most teams upload it to a cloud platform to measure, compare, and share. See our hub on scan-to-BIM to turn it into a model.
- View it in a browser, with nothing to install.
- Measure distances, areas, and volumes.
- Compare the as-built against a BIM model.
- Share it by link with clients and teams.
- Convert it into a mesh or a BIM model.
A point cloud is only as good as the capture. Good overlap and a clean site on the day save hours of processing later.
Create, then share with your clients
Once your cloud is ready, a link lets your client see it in 3D in the browser, with nothing to install.
To create a point cloud, pick a capture method (laser scanning, photogrammetry, mobile mapping, or a LiDAR-equipped iPhone Pro) based on precision, speed, and budget, then follow one workflow: capture, register, clean, export, and share. Good overlap and cleanup on site are what separate a usable cloud from a noisy one.
Frequently asked questions
Related articles
Point Cloud Definition: What It Is (3D, Simply)
A clear definition of a 3D point cloud: what it is, what it is made of, how it differs from a mesh and a BIM model, its types and formats.
Point Cloud File Formats: E57, LAS, LAZ, RCS, LGSx
The main point cloud file formats explained: E57, LAS, LAZ, RCS, RCP, LGSx. Open vs proprietary, which to use when, and how to convert.
Photogrammetry vs Laser Scanning: Which to Choose
Photogrammetry vs laser scanning compared: accuracy, cost, speed, conditions, and use cases, plus how to choose and when to combine both.