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

Unable to get free space #182

Open
4sfaloth opened this issue Jul 28, 2017 · 4 comments
Open

Unable to get free space #182

4sfaloth opened this issue Jul 28, 2017 · 4 comments
Labels

Comments

@4sfaloth
Copy link

I am facing this weird issue which I'm not sure if it's visualization (ie: octovis) related or not. I am generating occupancy maps using insertPointCloud function. Now, if I generate a very small map (eg: just 10 scans) then I can open it on Octovis and see both the occupied and the free space, and they seem to be correct. However, if I make a larger map (either by rotating in place or moving) then I can only visualize the occupied space in Octovis; free space appears to be empty.

Is there any known issue which could cause this?
Is there a way for me to check if the tree is being stored correct, to confirm if the issue is in octovis only?

@ahornung
Copy link
Member

ahornung commented Aug 4, 2017

This could be a visualization issue, maybe the number of free voxels is too large to display. Octovis displays the number of voxels. You could use that number to check if everything is integrated correctly, or write a small program that reads a .ot file and prints the statistics.

@peiyunh
Copy link

peiyunh commented Sep 21, 2019

I run into this issue as well. With 100k points, octovis fails to display free voxels. Once I cut it down to 20k, octovis starts cooperating. Does anyone by any chance know a workaround?

@ahornung
Copy link
Member

How many free nodes does your octree have? In contrast to the efficient octree encoding, the visualization needs a lot of memory for them.

The workaround is not to display the free space for large octree, or change the visualization resolution ("tree depth cutoff").
I'm pretty sure that the rendering can also be improved for memory efficiency, given someone wants to look deep into the code.

@peiyunh
Copy link

peiyunh commented Sep 26, 2019

Thanks for following up. I believe in my case there were more than 2 million free nodes when octovis stopped working. I ended up building another octree with a 2x larger voxel size just for the purpose of visualization.

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

No branches or pull requests

3 participants