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

High Performance Datastructures for Labeling #20

Open
maarzt opened this issue Mar 6, 2018 · 0 comments
Open

High Performance Datastructures for Labeling #20

maarzt opened this issue Mar 6, 2018 · 0 comments

Comments

@maarzt
Copy link
Collaborator

maarzt commented Mar 6, 2018

When creating dense labels on a big image 300x300x300 pixels:

  • drawing becomes slow
  • too much memory is used (12 bytes per labeled pixel)

We need better data structures to deal with these problems, possible approaches:

  • SparseCellImg
    • Special cell image for BitType, which needs less memory when a cell is completely black / white.
    • It has a cursor that skips the complete black cell or pixels, such that iterating over white pixels is fast.
  • Use OctTrees / NTrees
  • Caches
    • Deal with the high memory usage by applying disk caches.
  • Compression
    • Implement a compressed BitType. Compression maybe like LabelRegion, but with randomaccess and the possibility to change 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

1 participant