Normals from trajectory file - Hovermap SLAM data

Discuss CloudCompare software here.
Post Reply
Chani_ZynQ360
I have made <0 posts
I have made <0 posts
Posts: 2
Joined: Mon Jul 16, 2018 4:06 pm
5
Full Name: Chani Tough
Company Details: ZynQ 360 Limited
Company Position Title: Innovation Manager
Country: United Kingdom
Linkedin Profile: No

Normals from trajectory file - Hovermap SLAM data

Post by Chani_ZynQ360 »

I have been trying to create normals from an imported Hovermap .laz file but not getting much success at all. Is there a way in Cloudcompare to generate normals based on a trajectory file?

I have also tried in Metashape but it creates lots of inverted normals too. There is an option in Metahsape to use a trajectory file but I don't think it supports XYZ trajectory files.

I thought about trying to create TLS sensors along the trajectory file and using them to orient/create normals in the SLAM data but I don't see any option for that (or I'm not understanding the correct process).

Has anyone done anything like this before using Hovermap or other SLAM data?

I have put these questions on both the Cloudcompare forum and emailed Metashape support directly but neither of them have responded.
User avatar
Daniel Wujanz
V.I.P Member
V.I.P Member
Posts: 227
Joined: Fri Aug 24, 2018 11:26 am
5
Full Name: Daniel Wujanz
Company Details: technet GmbH
Company Position Title: 3D Laser Scanning Specialist
Country: Germany
Linkedin Profile: Yes
Location: Berlin
Has thanked: 206 times
Been thanked: 188 times
Contact:

Re: Normals from trajectory file - Hovermap SLAM data

Post by Daniel Wujanz »

Dear Chani,

if you have time stamps for both the trajectory and the point cloud then you can compute quasi static scans. You can then proceed with them as usual.

Have you got this information?

We have a prototype which does exactly what you need yet for another task. If you want, I can prepare the data for you.

See https://www.youtube.com/watch?v=1NN6r6juE7I&t=1908s at 34:52 min:sec

Cheers

Daniel
Chani_ZynQ360
I have made <0 posts
I have made <0 posts
Posts: 2
Joined: Mon Jul 16, 2018 4:06 pm
5
Full Name: Chani Tough
Company Details: ZynQ 360 Limited
Company Position Title: Innovation Manager
Country: United Kingdom
Linkedin Profile: No

Re: Normals from trajectory file - Hovermap SLAM data

Post by Chani_ZynQ360 »

Daniel Wujanz wrote: Tue Nov 01, 2022 3:05 pm Dear Chani,

if you have time stamps for both the trajectory and the point cloud then you can compute quasi static scans. You can then proceed with them as usual.

Have you got this information?

We have a prototype which does exactly what you need yet for another task. If you want, I can prepare the data for you.

See https://www.youtube.com/watch?v=1NN6r6juE7I&t=1908s at 34:52 min:sec

Cheers

Daniel
Hi Daniel, thanks for the reply. The Hovermap produces an xyz trajectory file which has the following columns: time gpstime x y z pitch roll yaw rot.w rot.x rot.y rot.z
The Hovermap itself stores the following point cloud attributes: Intensity, range, time, return number
(strongest & last) and ring number

I'm guessing that should be enough to match the time code of the points to the same time position in the trajectory.
User avatar
Daniel Wujanz
V.I.P Member
V.I.P Member
Posts: 227
Joined: Fri Aug 24, 2018 11:26 am
5
Full Name: Daniel Wujanz
Company Details: technet GmbH
Company Position Title: 3D Laser Scanning Specialist
Country: Germany
Linkedin Profile: Yes
Location: Berlin
Has thanked: 206 times
Been thanked: 188 times
Contact:

Re: Normals from trajectory file - Hovermap SLAM data

Post by Daniel Wujanz »

Nice, could you send me a chunk of that data to see if we can do something with it or is it critical data? If the latter applies, is there a demo dataset that I could use?

Cheers

Daniel
User avatar
RobM
I have made 100> posts
I have made 100> posts
Posts: 101
Joined: Mon Aug 22, 2016 9:55 am
7
Full Name: Robin Miller
Company Details: My own boss
Company Position Title: Managing Director
Country: Germany
Linkedin Profile: No
Been thanked: 14 times

Re: Normals from trajectory file - Hovermap SLAM data

Post by RobM »

I just looked at your video.
as we talked about the time stamp topic before, one thing came to my mind:
with the VLX, wouldnt it be sufficient to cut the "point sausage" ,as you called it, exactly at those points where panoramic images are taken? because there are known timestamps for the images and I`d assume with the image being present there should be an option to turn them into "pseudo structured scans". in your presentation you wrote "where static behavior can be assumed" and I am thinking, as you stop to take the image with the VLX that might be exactly what you are looking for?

Daniel Wujanz wrote: Tue Nov 01, 2022 3:05 pm Dear Chani,

if you have time stamps for both the trajectory and the point cloud then you can compute quasi static scans. You can then proceed with them as usual.

Have you got this information?

We have a prototype which does exactly what you need yet for another task. If you want, I can prepare the data for you.

See https://www.youtube.com/watch?v=1NN6r6juE7I&t=1908s at 34:52 min:sec

Cheers

Daniel
User avatar
MarkJLane
V.I.P Member
V.I.P Member
Posts: 149
Joined: Sat Mar 17, 2018 11:25 pm
6
Full Name: Mark Lane
Company Details: Faro Technologies
Company Position Title: ANZ Account Manager
Country: Australia
Linkedin Profile: Yes
Has thanked: 25 times
Been thanked: 43 times
Contact:

Re: Normals from trajectory file - Hovermap SLAM data

Post by MarkJLane »

Chani_ZynQ360 wrote: Tue Nov 01, 2022 1:35 pm Has anyone done anything like this before using Hovermap or other SLAM data?
Using Cloud Compare, you should be able to save that trajectory file into a format that Metashape will accept.

Alternatively, with the ZEB Horizon and Vision camera, I can export a structured E57 based on the camera interval. Using this E57 I can calculate normals based on sensor origin.
Mark Lane
Sydney, Australia
TLS Tools
GeoSLAM
Image
User avatar
Daniel Wujanz
V.I.P Member
V.I.P Member
Posts: 227
Joined: Fri Aug 24, 2018 11:26 am
5
Full Name: Daniel Wujanz
Company Details: technet GmbH
Company Position Title: 3D Laser Scanning Specialist
Country: Germany
Linkedin Profile: Yes
Location: Berlin
Has thanked: 206 times
Been thanked: 188 times
Contact:

Re: Normals from trajectory file - Hovermap SLAM data

Post by Daniel Wujanz »

RobM wrote: Wed Nov 30, 2022 9:03 pm I just looked at your video.
as we talked about the time stamp topic before, one thing came to my mind:
with the VLX, wouldnt it be sufficient to cut the "point sausage" ,as you called it, exactly at those points where panoramic images are taken? because there are known timestamps for the images and I`d assume with the image being present there should be an option to turn them into "pseudo structured scans". in your presentation you wrote "where static behavior can be assumed" and I am thinking, as you stop to take the image with the VLX that might be exactly what you are looking for?
Dear Robin,

I've had many discussions with Navvis at Intergeo and the conclusion is (I cannot say more) that the system operates slightly different than others. The problem is that timestamps for single points cannot be retrieved. Hence, you cannot associate single points to image timestamps.

However, we're in close exchange with Navvis to find a solution.

All the best

Daniel
Post Reply

Return to “CloudCompare”