Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

use offline_viewer to registrate multi-view point clouds #125

Open
CharlieV5 opened this issue Dec 5, 2024 · 6 comments
Open

use offline_viewer to registrate multi-view point clouds #125

CharlieV5 opened this issue Dec 5, 2024 · 6 comments

Comments

@CharlieV5
Copy link

Hi, Koide-san!
I am impressed with the effect of global optimization with multiple loop constraints. And I want to test if the multi-view TLS( Terrestrial Laser Scanner ) point clouds can be registated well by GLIM offline_viewer.
So I write some code to make the dump data for offline_viewer as follow.
test_global_mapping.zip
test-data.zip
The test-data contains 8 stations of TLS point cloud, and the pose of each station. The test_global_mapping will convert the test-data into dump data for offline_viewer. In the code, I convert the xyz clouds into bin files and construct the factor graph for the multiple stations.
Then I get the dump data, the offline_viewer can load the data and show the right multi-view point clouds. But after choosing optimization and finding overlapping submap, it crashed.

Any advice to make it possible to registrate multi-view cloud using GLIM?

20241205_114906.mp4
@koide3
Copy link
Owner

koide3 commented Dec 29, 2024

Wow, it's an interesting trial. I guess we need an additional prior factor to fix the gauge freedom of the entire graph. If you could share the dump file, I would like to try implementing a mechanism to detect and fix the exception.

@CharlieV5
Copy link
Author

Wow, it's an interesting trial. I guess we need an additional prior factor to fix the gauge freedom of the entire graph. If you could share the dump file, I would like to try implementing a mechanism to detect and fix the exception.

OK,this is the dump file for you.
https://drive.google.com/file/d/1GeabUtvAkszC1CVOchTDegXcK1r-NoAx/view?usp=drive_link

@koide3
Copy link
Owner

koide3 commented Jan 8, 2025

I found that the error is caused by submap point cloud data that contains corrupted covariance matrices. I just pushed a patch to detect and fix corrupted point cloud #131. I confirmed your dump file can be successfully loaded and optimized with this patch.

@CharlieV5
Copy link
Author

Thank you, Koide-san! I've tested the latest code with the dump data. It works great! I will test with more TLS data later.

@koide3
Copy link
Owner

koide3 commented Jan 10, 2025

Your welcome! BTW, feeding TLS data to GLIM would be of interest to people. Do you mind if I share the mapping result of your dump data on SNS (X)?

@CharlieV5
Copy link
Author

Your welcome! BTW, feeding TLS data to GLIM would be of interest to people. Do you mind if I share the mapping result of your dump data on SNS (X)?

Not at all. Feel free to share the data!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants