Skip to content

Superpixel Pooling implemented in PyTorch and Chainer

License

Notifications You must be signed in to change notification settings

WxWstranger/superpixPool

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

superpixPool

Superpixel Pooling implemented in PyTorch and Chainer

illustration

This code is the product of a Masters thesis done by Mathijs Schuurmans under supervision by Maxim Berman and Matthew Blaschko, at the center for processing speech and images, Dept. of Electrical Engineering, KU Leuven. It includes CUDA implementation of superpixel pooling with integration in the Chainer and PyTorch deep learning framworks.

Experimental details related to this work are given in https://arxiv.org/abs/1806.02705. Please cite that article if you use this code in your research.

At the moment the chainer version implements a 3d pooling layer, and the pytorch version a 2d pooling layer. See the PyTorch and Chainer subfolders for implementation of the pooling layer in these respective frameworks.

WARNING: we are not actively maintaining this package. Hopefully it can be useful to some (perhaps as a starting point) but it is difficult for us to keep up with ensuring compatibility among compilers/python/pytorch versions.

About

Superpixel Pooling implemented in PyTorch and Chainer

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 73.0%
  • Cuda 23.8%
  • C++ 3.2%