https://github.com/qq456cvb/PRIN

Pointwise Rotation-Invariant Network (AAAI 2020)
https://github.com/qq456cvb/PRIN

pytorch rotation-invariant-features segmentation

Last synced: about 2 months ago
JSON representation

Pointwise Rotation-Invariant Network (AAAI 2020)

• Host: GitHub
• URL: https://github.com/qq456cvb/PRIN
• Owner: qq456cvb
• Created: 2018-11-21T08:52:54.000Z (almost 6 years ago)
• Default Branch: master
• Last Pushed: 2021-02-25T09:12:01.000Z (over 3 years ago)
• Last Synced: 2024-07-22T03:37:16.359Z (2 months ago)
• Topics: pytorch, rotation-invariant-features, segmentation
• Language: Python
• Homepage:
• Size: 23.4 KB
• Stars: 85
• Watchers: 7
• Forks: 10
• Open Issues: 1
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# PRIN

## Pointwise Rotation-Invariant Network in PyTorch

# News

An improved version of PRIN (SPRIN) is released [here](https://github.com/qq456cvb/SPRIN) and described in [PRIN/SPRIN: On Extracting Point-wise Rotation Invariant Features](https://arxiv.org/abs/2102.12093), which achieves much better results.

## Overview

This repository is the Pytorch implementation of [PRIN (Pointwise Rotation-Invariant Network)](https://arxiv.org/pdf/1811.09361.pdf).

## Dependencies

* Install s2cnn (https://github.com/jonas-koehler/s2cnn) and its dependencies (pytorch, cupy, lie_learn, pynvrtc).

* Install pybind11 and compile the script under src (https://pybind11.readthedocs.io/)

## Dataset and pretrained weights

* Download ShapeNet 17 Part Segmentation Dataset in h5py format from 

https://drive.google.com/drive/folders/1wC-DpeRtxuuEvffubWdhwoGXGeW052Vy?usp=sharing

* Download pretrained weights (trained on unrotated shapes) from

https://drive.google.com/open?id=1QnFqQdWmx0cYtYeN9tJNlf-E5ZLawRBv

## Usage

* For training, run "python train.py --log_dir log --model_path ./model.py --num_workers 4"

* For testing, run "python test.py --weight_path ./state.pkl --model_path ./model.py --num_workers 4"

## License

MIT

## References

Our paper is available on https://arxiv.org/abs/1811.09361.

## Citation

@inproceedings{you2020pointwise,  

  title={Pointwise rotation-invariant network with adaptive sampling and 3d spherical voxel convolution},  

  author={You, Yang and Lou, Yujing and Liu, Qi and Tai, Yu-Wing and Ma, Lizhuang and Lu, Cewu and Wang, Weiming},  

  booktitle={Proceedings of the AAAI Conference on Artificial Intelligence},  

  volume={34},  

  number={07},  

  pages={12717--12724},  

  year={2020}  

}