Skip to content

Reimplementation "PointGroup: Dual-Set Point Grouping for 3D Instance Segmentation (CVPR2020)"

Notifications You must be signed in to change notification settings

K-nowing/PointGroup-PyTorch

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

PointGroup-PyTorch

Reimplementation of PointGroup with different backbone(KPConv, DGCNN) for study purposes.

Reimplementation Process Post

Model Architecture

I used KPConv for Backbone, DGCNN for ScoreNet

Inference Result

Semantic segmentation

Instance segmentation

Training

Simply run the following script to start the training:

    python train.py

Testing

Simply run the following script to start the testing(recommended to run only 1 epoch):

    python test.py   # you should change weight directory

Reference

  1. KPConv-PyTorch
  2. dgcnn.pytorch
  3. visualizing with pptk

About

Reimplementation "PointGroup: Dual-Set Point Grouping for 3D Instance Segmentation (CVPR2020)"

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published