Skip to content

sethdusek/kmeanswgpu

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 
 
 
 
 

Repository files navigation

K means clustering for images

Usage

  cargo run <path to image> <k>

Example:

Input Image Output (k = 50)
image image

Limitations

  • Arbitrary n-dimensional data isn't supported and isn't planned to be supported. This is only really meant for image segmentation
  • No convergence tracking yet, currently the number of k-means iterations is hardcoded
  • K (#clusters) must be specified manually at the moment

Planned

  • Add support for reading-out results from the GPU. Currently the results can only be viewed visually
  • Auto-select k instead of specifying it manually, using the Elbow Method or similar
  • Convergence tracking

About

GPU-accelerated k-means clustering for images

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published