Mesh from imported las file

I imported a colorized las file from a mobile Lidar and would like to process a mesh, however mesh is grayed out and not able to be selected. I’ve read that Pix4Dmatic is capable of creating a mesh from a las file without any images, but cannot find a way to make mesh selectable under processing tab. Any thoughts would be appreciated. Thank you.

Hello,

Thank you for reaching out.

For generating a textured mesh in PIX4Dmatic, the process requires the images associated with the point cloud.

Could you please let me know which LiDAR model you are using? If the scanner also captures images, they can be imported directly into the software to create the textured mesh.

Kind regards,
Alexey

Hello Alexey,
Thank you for your response.
The mobile LiDAR we use does have image collection. The model is the Foxtech SLAM100 Handheld Imaging LiDAR Scanner.
Thank you,
Joe

Hello Joe,

Thank you for your reply.

To generate a high-quality mesh in PIX4Dmatic from your LiDAR point cloud, I recommend the following workflow:

  1. Import the images captured by the scanner.
  2. Import the LiDAR point cloud and the corresponding scanner trajectory file.
  3. Run Calibration. In the processing options, set the Externals Confidence parameter to ‘Absolute’. This is a key step to ensure the images remain perfectly aligned with the LiDAR data.
  4. Run the Mesh step to generate the final model.

I hope this workflow helps you achieve the desired results. Please let me know if you have any further questions.

Kind regards,

Alexey

Hello Alexey,
Thank you for this information. I am currently working on this process and ran into something that I could use assistance with.

That data I am working with is from a mobile LiDAR unit, FoxTech SLAM 100. The images have no data associated with them and for some reason were captured rotated 90degrees to the right as shown in the attached image.

I have the camera position file which I have used and ran the calibration successfully with 100% calibrated, but the images still have not oriented themselves correctly (see below). I also tried rotating the images prior to importing them with no success. As you see the images show up correctly in the viewing pane but not in the model area. Is there something I am missing? I’d appreciate any insight you may have.

(Attachment cam0_pic271_Monday-11-57-31-574522.jpeg is missing)

(Attachment camera_pos.csv is missing)

(Attachment cam0_pic271_Monday-11-57-31-574522.jpeg is missing)

(Attachment camera_pos.csv is missing)

Hello Alexey,
Thank you for this information. I am currently working on this process and ran into something that I could use assistance with.

That data I am working with is from a mobile LiDAR unit, FoxTech SLAM 100. The images have no data associated with them and for some reason were captured rotated right 90degrees.

I have the camera position file which I have used and ran the calibration successfully with 100% calibrated, but the images still have not oriented themselves correctly (see below). I also tried rotating the images prior to importing them with no success. As you see the images show up correctly in the viewing pane but not in the model area. Is there something I am missing? I’d appreciate any insight you may have.

Thank you,
Joe

Hello Joe,

Thank you for your message and for providing the screenshot.

It appears the images are not being oriented correctly after the calibration step as we can see in the 3D view. This can happen when the flight path is a single straight line and the images are missing accurate orientation data.

To resolve this, you could try one of the following:

  • Capture images along a more complex path (e.g., a grid pattern).
  • Add ground control points to your project to strengthen the geometry.

Hope this helps.

Kind regards,

Alexey

Thank you, Alexey.
We were attempting something a bit different than I am used to with Pix4D. I have processed many drone flights. However, this was not a drone flight but rather a mobile LiDAR unit carried on a backpack. So, it was just single passes between interior building storage racks. I had concerns about the overlap and also noticed that the images captured from the LiDAR unit are only 5mp. I’d imagine that doesn’t help as I read that the minimum mp for Pix4D is 14. Can you please confirm the mp requirement?

Thank you for your response and confirming my initial thoughts.

-Joe

Hello Joe,

I think you can still try running the calibration with the 5MB image size. The main challenge with the dataset you shared is the trajectory. If all the images are captured along a single straight line and there is no orientation information, the software cannot reliably find the correct orientation values.

Having a more complex trajectory, where you see the same location from different images not lying on the same line, should help to establish the correct orientation.

Let me know how it goes.

Kind regards,

Alexey

Thank you for your insight, Alexey.
Take care,
Joe