• Dear visitors,

    The email issue seems to have been finally solved.
    Thank you for your patience and happy browsing.

    Team ACM.

TUTORIAL LIDAR (Point cloud) to mesh tutorial

LilSKi

Well-Known Member
Should be able to assign the same center point value to all the tiles you add in. Just write the one down from the center most tile and use that number for all the tiles after that.
 

AkinOstrich

New Member
Hello everybody! I'm remaking my neighborhood in Blender. I used some LIDAR data, trimmed lots of vertices, and exported the .fbx file from Blender. However, when I tried to open the .fbx file in AC Editor, it had trouble importing (meaning no progress loading at all). Has anybody else had this problem?

Just trying to drive with basic Blender textures. Using checkered for terrain and brick for all boxes. I have respected all naming conventions, but still not sure what's going on. Any tips?

EDIT: The max amount of vertices for one mesh supported by AC Editor is 65536. I had 241,721.
 
Last edited:

maruto

Active Member
Hello everybody! I'm remaking my neighborhood in Blender. I used some LIDAR data, trimmed lots of vertices, and exported the .fbx file from Blender. However, when I tried to open the .fbx file in AC Editor, it had trouble importing (meaning no progress loading at all). Has anybody else had this problem?

Just trying to drive with basic Blender textures. Using checkered for terrain and brick for all boxes. I have respected all naming conventions, but still not sure what's going on. Any tips?

EDIT: The max amount of vertices for one mesh supported by AC Editor is 65536. I had 241,721.

Hello @AkinOstrich ,

yeah you have to make sure you split large meshes like terrain and road to smaller section so it never goes higher than 65k vertices....

welcome here and share your work if you feel like it, I'm sure many of us will enjoy discovering your neigbourhood! (and help you bringing it to AC)
 

AkinOstrich

New Member
Hi guys! I've found out that some users might need to select "Number of Returns" in the Scalar Fields Active drop-down box. It wasn't working too well for me when I used "Classification". This may be because of the source of the LIDAR data? Anyway, the sliders in the SF display params worked well separating the ground from the trees after that.

From there, as @DanTDBV said, use "Filter Points By Value", uncheck the sample which you want gone, and convert to mesh (Delaunay 2.5D (XY Plane)). Thanks guys!

upload_2020-12-19_16-21-32.png

Filter Points By Value tool

upload_2020-12-19_16-9-20.png

Number of Returns (right slider under 2)

upload_2020-12-19_16-10-44.png

Classification (right slider under 2)
 

Mr Whippy

Active Member
I found this earlier, while looking for where I'd originally downloaded the environment agency LAZ files.

https://houseprices.io/lab/lidar/map

It's a great way to just find the coverage of the open LAZ scans, zoom in, look around, and then generate a mesh from it to ASC format, and save it.
However, I know these scans are best as point clouds, the mesh generator 'drapes' a cloth over the scene, so data stuff like roads under trees is completely lost.

Also I'm not sure if it's using the densest point cloud data, or just a lower quality version.

I'm unsure if it's up to date with latest etc, but it seems quite accurate to what I had found in 2017 ish)


In any case, a quick handy tool to figure out what data you might want, grab it, and then get the better data if needed later!
 

Mr Whippy

Active Member
Wow, that's amazing, colour too!

Possibly posted earlier, same resource as I was using in 2017 earlier in this thread.

https://environment.data.gov.uk/DefraDataDownload/?Mode=survey



But seemingly they did a national LIDAR survey in 2019 and from what I can see coverage is either the same or better than the previous surveys where I've looked.

Definitely worth checking over project areas you're interested in, grabbing all the relevant LIDAR sets, and merging them all in Cloud Compare to see what you've got, if you were disappointed before.



I'm going to see how viable it is to use the raster ortho photos to colourise the point clouds via Cloud Compare.


Cheers
 
Top