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

artifacts due to not connected triangles #7

Open
tlow92 opened this issue Mar 10, 2020 · 3 comments
Open

artifacts due to not connected triangles #7

tlow92 opened this issue Mar 10, 2020 · 3 comments
Assignees

Comments

@tlow92
Copy link

tlow92 commented Mar 10, 2020

Hi,

I am trying to reconstruct a mesh from a point cloud that was exported using google cartographers assets writer. The data used is from "Deutsche Museum" which was scanned by the google cartographer team.

I played around with different parameters such as the smallRegion, fillHoles, ransac, and the different algorithms MC, PMC, SF and MT.
No matter what parameters I choose, I always have alot of holes in the mesh, which look like the triangles are not connected correctly.

In this dataset the robot has a fixed position. I understand that since I don't provide the position of the robot (which is 0, 0, 0) the orientation of the normals can be inconsistent. Is this a problem for LVR when connecting the triangles?

Could it be, that the quality of the point cloud is just not good enough? Its a single room, which consists of 50M points.

Any advice is appreciated.

image

@GPrathap
Copy link

GPrathap commented Jun 2, 2022

I also have similar issue,
Screenshot from 2022-06-02 15-47-42

@twiemann
Copy link
Contributor

twiemann commented Jun 3, 2022

Hi GPrathap, it seems the voxel size you are using is tool small, i.e., smaller then the point density in your cloud. Try to increase the -v parameter by a factor of 10 or larger. To get consistent surfaces you might also consider the number of nearest neighbor used for normal estimation and signed distance computation. This will come with larger runtimes but better approximations

@GPrathap
Copy link

GPrathap commented Jun 3, 2022

@twiemann Yeap it works thank youu

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

4 participants