Skip to content

The code is for the paper "Rethinking and Improving Few-Shot Segmentation from a Contour-Aware Perspective"

Notifications You must be signed in to change notification settings

hardtogetA/CTANet

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

2 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

CTANet

The code is for the paper "Rethinking and Improving Few-Shot Segmentation from a Contour-Aware Perspective"

Overview

  • COCO/ includes scripts for COCO-20i dataset
  • FSS/ includes scripts for FSS-1000 dataset
  • VOC/ includes scripts for Pascal-5i dataset

And for each folder:

  • model/ includes related models and modules
  • store/ includes pretrained models under 1-shot setting
  • others/ includes script of loss functions

Get started

Requirements

python == 3.9, torch == 1.9, numpy, einops, opencv-python

Dataset

Related datasets Pascal-5i / COCO-20i / FSS-1000 can be found in Google Driver

Test and Train

  • Specify the path of datasets in get_data.py
  • Specify the path of pre-trained models in the test file
  • Use the following command
    python Test.py/train.py
    

About

The code is for the paper "Rethinking and Improving Few-Shot Segmentation from a Contour-Aware Perspective"

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages