No description, website, or topics provided.
Switch branches/tags
Nothing to show
Clone or download
Latest commit d767b0e Mar 24, 2018
Permalink
Failed to load latest commit information.
misc update readme Feb 10, 2018
models Add part segmentation Feb 4, 2018
part_seg Update and rename Readme.md to README.md Feb 5, 2018
sem_seg Update README.md Feb 5, 2018
utils Update tf_util.py Feb 12, 2018
.gitignore add gitignore Feb 3, 2018
README.md update readme Feb 10, 2018
evaluate.py add evaluation Mar 24, 2018
provider.py Add part segmentation Feb 4, 2018
train.py add edge conv Feb 3, 2018

README.md

Dynamic Graph CNN for Learning on Point Clouds

We propose a new neural network module dubbed EdgeConv suitable for CNN-based high-level tasks on point clouds including classification and segmentation. EdgeConv is differentiable and can be plugged into existing architectures.

[Project] [Paper]

Overview

DGCNN is the author's re-implementation of Dynamic Graph CNN, which achieves state-of-the-art performance on point-cloud-related high-level tasks including category classification, semantic segmentation and part segmentation.

Further information please contact Yue Wang and Yongbin Sun.

Requirements

Point Cloud Classification

  • Run the training script:
python train.py
  • Run the evaluation script after training finished:
python evalutate.py

Citation

Please cite this paper if you want to use it in your work,

@article{dgcnn,
  title={Dynamic Graph CNN for Learning on Point Clouds},
  author={Yue Wang, Yongbin Sun, Ziwei Liu, Sanjay E. Sarma, Michael M. Bronstein, Justin M. Solomon},
  journal={arXiv preprint arXiv:1801.07829},
  year={2018}
}

License

MIT License

Acknowledgement

This code is heavily borrowed from PointNet.