Skip to content

[CVPR 2021] CGA-Net: Category Guided Aggregation for Point Cloud Semantic Segmentation

Notifications You must be signed in to change notification settings

MCG-NJU/CGA-Net

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

33 Commits
 
 
 
 
 
 

Repository files navigation

CGA-Net: Category Guided Aggregation for Point Cloud Semantic Segmentation

by Lu Tao, Wang Limin

@inproceedings{lu2021cga,
  title={CGA-Net: Category Guided Aggregation for Point Cloud Semantic Segmentation},
  author={Lu, Tao and Wang, Limin and Wu, Gangshan},
  booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  pages={11693--11702},
  year={2021}
}

Introduction

This is the official implementation of CGA-Net: Category Guided Aggregation for Point Cloud Semantic Segmentation, which proposes to utilize different aggregation strategies between the same category and different categories. Here we provide the Tensorflow version. The code is very clear and easy to transplant to other frameworks.

Usage

This module can be leveraged in any existing point-based segmentation networks. Here we provide an example of how to apply CGA module to CloserLook3D, please refer to examples/CloserLook3D/README.md.

For other backbones, one can try to modify the source code in CGA/cga.py.

Acknowledgement

Our Tensorflow code is based on CloserLook3D, RandLA-Net, MeteorNet and we benefit a lot from PointNet2, KPConv.

About

[CVPR 2021] CGA-Net: Category Guided Aggregation for Point Cloud Semantic Segmentation

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages