https://github.com/faultaddr/rscnn
https://github.com/faultaddr/rscnn
Last synced: 8 months ago
JSON representation
- Host: GitHub
- URL: https://github.com/faultaddr/rscnn
- Owner: faultaddr
- Created: 2020-09-22T06:35:52.000Z (over 5 years ago)
- Default Branch: master
- Last Pushed: 2020-09-26T21:23:36.000Z (over 5 years ago)
- Last Synced: 2025-03-25T21:22:07.078Z (about 1 year ago)
- Language: Python
- Size: 10.5 MB
- Stars: 7
- Watchers: 2
- Forks: 4
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
Relation-Shape Convolutional Neural Network for Point Cloud Analysis
===
This repository contains the author's implementation in Pytorch for the paper:
__Relation-Shape Convolutional Neural Network for Point Cloud Analysis__ [[arXiv](https://arxiv.org/abs/1904.07601)] [[CVF](http://openaccess.thecvf.com/content_CVPR_2019/papers/Liu_Relation-Shape_Convolutional_Neural_Network_for_Point_Cloud_Analysis_CVPR_2019_paper.pdf)]
[Yongcheng Liu](https://yochengliu.github.io/), [Bin Fan](http://www.nlpr.ia.ac.cn/fanbin/), [Shiming Xiang](https://scholar.google.com/citations?user=0ggsACEAAAAJ&hl=zh-CN) and [Chunhong Pan](http://people.ucas.ac.cn/~0005314)
[__CVPR 2019 Oral & Best paper finalist__](http://cvpr2019.thecvf.com/) __Project Page__: [https://yochengliu.github.io/Relation-Shape-CNN/](https://yochengliu.github.io/Relation-Shape-CNN/)
## Citation
If Liu's paper is helpful for your research, please consider citing:
```BibTex
@inproceedings{liu2019rscnn,
author = {Yongcheng Liu and
Bin Fan and
Shiming Xiang and
Chunhong Pan},
title = {Relation-Shape Convolutional Neural Network for Point Cloud Analysis},
booktitle = {IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
pages = {8895--8904},
year = {2019}
}
```
## Usage: Preparation
## Note
This repository is reproduction of RSCNN based on Pytorch version 1.0 or 1.1, borrowed from Liu's code (https://yochengliu.github.io/Relation-Shape-CNN/)
### Requirement
- Ubuntu 16.04
- Python 3.6 (recommend Anaconda3)
- Pytorch 1.0+ (test on 1.0, 1.1 and 1.2)
- CMake > 2.8
- CUDA 10.2 + cuDNN 7.6
- GNU <= 7.5
Setup
-----
The custom ops used by Pointnet++ are currently **ONLY** supported on the GPU using CUDA.
* Building `_ext` module (build and copy a .so file to utils/)
::
python setup.py build_ext --inplace
### Dataset
__Shape Classification__
Download and unzip [ModelNet40](https://shapenet.cs.stanford.edu/media/modelnet40_ply_hdf5_2048.zip) (415M). Replace `$data_root$` in `cfgs/config_*_cls.yaml` with the dataset parent path.
__ShapeNet Part Segmentation__
Download and unzip [ShapeNet Part](https://shapenet.cs.stanford.edu/media/shapenetcore_partanno_segmentation_benchmark_v0_normal.zip) (674M). Replace `$data_root$` in `cfgs/config_*_partseg.yaml` with the dataset path.
## Usage: Training
### Shape Classification
sh train_cls.sh
You can modify `relation_prior` in `cfgs/config_*_cls.yaml`. We have trained a Single-Scale-Neighborhood classification model in `cls` folder, whose accuracy is 92.38%.
### Shape Part Segmentation
sh train_partseg.sh
We have trained a Multi-Scale-Neighborhood part segmentation model in `seg` folder, whose class mIoU and instance mIoU is 84.18% and 85.81% respectively.
## Usage: Evaluation
### Shape Classification
Voting script: voting_evaluate_cls.py
You can use our model `cls/model_cls_ssn_iter_16218_acc_0.923825.pth` as the checkpoint in `config_ssn_cls.yaml`, and after this voting you will get an accuracy of 92.71% if all things go right.
### Shape Part Segmentation
Voting script: voting_evaluate_partseg.py
You can use our model `seg/model_seg_msn_iter_57585_ins_0.858054_cls_0.841787.pth` as the checkpoint in `config_msn_partseg.yaml`.
## License
The code is released under MIT License (see LICENSE file for details).
## Acknowledgement
The code is heavily borrowed from [Pointnet2_PyTorch](https://github.com/erikwijmans/Pointnet2_PyTorch).
## Contact
If you have some ideas or questions about our research to share with us, please contact