Purchase orders accepted from universities, colleges and government agencies. Financing available with Affirm. Financing with Affirm. W-9 and vendor pack +1 888 850 6533
MySurveyingDirect
0
Home/Sample capture data

Sample capture data

Every point cloud on this site came off an instrument we sell. No renders posing as scans, no borrowed datasets, and no marketing department's idea of what a scan looks like. These were captured on a UK university campus with a handheld SLAM scanner, walked rather than flown. Download a preview now, with no form and no email address, or ask for the full datasets.

Download

Campus exterior, preview 525,339 points, decimated evenly from 40.5 million. 148 m by 187 m of courtyard, path and building faces. 17.0 MB · LAS 1.2 Interior, preview 526,369 points, decimated evenly from 28.4 million. 30 m by 37 m, captured with no satellite signal and no control network. 17.1 MB · LAS 1.2 Campus exterior, solved trajectory 6,873 positions. The route the operator actually walked, as the scanner solved it. 228 KB · LAS 1.2 Interior, solved trajectory 3,314 positions. Worth opening alongside the cloud to see where density comes from. 110 KB · LAS 1.2

The previews are a uniform decimation, every 77th point of the exterior and every 54th of the interior, taken across the whole file rather than truncated. That thins the cloud evenly over the entire walk, so what you are judging is the same capture at lower density rather than a selected corner of it. Nothing was cleaned, filtered or manually tidied.

Want the full files? The exterior is 1.3 GB and the interior is 922 MB, which is past what we can serve from a product page. Email support@mysurveyingdirect.com with the word "datasets" and we will send a link. No form gate, no drip sequence and no sales call unless you ask for one.

What is in the files

Captured 27 May 2026 on a UK university campus with an FJ Dynamics Trion handheld SLAM scanner.
  Campus exterior Interior
Full dataset 1.3 GB, 40,451,057 points 922 MB, 28,423,912 points
Preview above 17.0 MB, 525,339 points 17.1 MB, 526,369 points
Format LAS 1.2, point data format 3 LAS 1.2, point data format 3
Extent 148 m by 187 m, 52 m of height 30 m by 37 m, 6 m of height
Method Walked loop, handheld SLAM Walked, handheld SLAM
Conditions Outdoors, ordinary May daylight Satellite denied, no control network, no targets
Color in the file Elevation ramp, not photographic RGB. See the note below Elevation ramp, not photographic RGB
Relative accuracy Under 1 cm, post processed Under 1 cm, post processed
Also supplied Solved trajectory Solved trajectory
One thing worth being straight about. The renders further down this page are RGB colorized, taken in the viewer, and they are what the capture looks like in true color. The LAS files you can download above are not RGB colorized: their color fields carry an elevation ramp instead, which is how this particular export was written. We checked rather than assumed, and the numbers are unambiguous: 593 distinct colors across half a million points, and color correlating with height at 0.99. If you were expecting photographic color in the download, that is why it is not there. Ask and we will export an RGB colorized version.

What the output looks like

RGB colorized point cloud of a university campus courtyard, showing red brick buildings, a tree and a tarmac path in true color
Campus exterior, RGB colorized in the viewer. Brick, foliage and tarmac all read as themselves, which is the test of a colorized capture.
Plan view of the outdoor point cloud showing the walked survey trajectory across the campus
The same capture in plan, showing the route walked. Coverage is a function of where you went, which is the thing to understand about SLAM before you buy one.
Outdoor point cloud colored by height, showing terrain and building elevation variation
Colored by height. This is the coloring the downloadable LAS carries, so it is the fair comparison to what you will open.
Interior point cloud colored by height, captured with no satellite signal and no control network
Interior, colored by height. No satellite signal, no control network, no targets. This is the case that separates SLAM from everything else.

How to judge a point cloud you have been sent

Most sample data published by equipment vendors is chosen to flatter the instrument. Here is what to actually look at, including on ours.

  • Open it at full density, not decimated. The previews above are decimated and we say so on the button. If a vendor will only show you their data inside their own viewer, that is the finding.
  • Look at the edges of things. Building corners, kerbs, the tops of walls. Noise shows up at edges long before it shows up on a flat surface.
  • Follow the trajectory back to the start. On a walked loop the beginning and the end are the same place. How far apart are they in the cloud? That is drift, and it is the honest measure of a SLAM instrument. The trajectory files above are supplied so you can do exactly this.
  • Check the featureless stretches. Long corridors, blank walls and open fields are where SLAM has nothing to track against. Look at those specifically rather than at the interesting geometry.
  • Ask what was cleaned. If a dataset has had its noise manually removed before publication, it is a picture of somebody's afternoon rather than of the instrument. Ours has not been.
  • Read the header, not the marketing. A LAS header states the point count, the scale factors and the bounding box. If they disagree with what you were told, believe the header.
On accuracy figures generally. Under 1 cm relative, post processed, is what these captures hold. Relative accuracy is not absolute accuracy: it describes how well the cloud agrees with itself, not how well it agrees with a national grid. A specification quoting one number without saying which, and without the conditions it was measured under, is not telling you anything. That applies to our figures as much as anyone's, which is why the conditions are in the table above. More on this on how we specify.

What was captured, and what was not

These are real working captures rather than a demonstration staged for a camera. The site is a UK university campus, walked in ordinary conditions in May. We do not name the institution, because a capture belongs to the people whose site it is and publishing the name is not ours to do.

If you want to see the instrument against something closer to your own problem, say what that is. A quarry, a corridor, a plant room, a heritage interior. We would rather show you a capture that answers your question than a prettier one that does not.

Ask for the full datasets, or for something closer to your work

The previews above need no email address. For the full 1.3 GB and 922 MB files, or an RGB colorized export, just ask.

Email support@mysurveyingdirect.com
Call +1 888 850 6533, Monday to Friday, 9 to 5 Mountain Time

We reply no later than the next working day. If you do not have software to open a LAS file, CloudCompare is free and will open all four.

Deciding between instruments rather than judging one? Compare instruments sets the handheld range side by side with the conditions attached, including the cases where the cheaper instrument is the right one. Complete systems shows what a working configuration costs.