Skip to content
spatial transformer for 3d point clouds
Python Shell
Branch: master
Clone or download
Latest commit 3fb388a Oct 7, 2019
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
assets sampl-partseg Oct 7, 2019
caffe-deform sampl-partseg Oct 7, 2019
sampling-based sampl-partseg Oct 7, 2019
.DS_Store update caffe Oct 7, 2019
.gitignore sampl Oct 4, 2019
README.md update caffe Oct 7, 2019

README.md

Spatial Transformer for 3D Point Clouds

[Project] [Paper]

Overview

This is the author's re-implementation of the long-tail recognizer described in:
"Spatial Transformer for 3D Point Clouds"
Jiayun WangRudrasis ChakrabortyStella X. Yu  (UC Berkeley / ICSI)  in arXiv 2019

Further information please contact Jiayun Wang.

Update notifications

  • 10/07/2019: Uploaded sampling-based methods for ShapeNet part segmentation.

Requirements

  • PyTorch (for the point-based method, version >= 0.4.1)
  • CAFFE (for the sampling-based method, please use our version as we rewrite some source codes.)
  • NCCL (for multi-gpu in the sampling-based method)

Sampling-based Methods

Install Caffe

Please use our version of CAFFE, as we provide the implementation of spatial transformers for bilateralNN, as described in the paper. A guide to CAFFE installation can be found here.

Data Preparation

See instructions in data/README.md.

Running Examples

* ShapeNet Part segmentation
    * train and evaluate
        ```bash
        cd sampling-based/exp/shapenet3d
        ./train_test.sh
    * test trained model
        ```bash
        cd sampling-based/exp/shapenet3d
        ./test_only.sh
        ```
        Predictions are under `pred/`, with evaluation results in `test.log`.

Benchmarks and Model Zoo

Please refer to Section 4 of the paper.

CAUTION

The code is implemented based on Dynamic Graph CNN, BilateralNN and SplatNet.

License and Citation

The use of this software is RESTRICTED to non-commercial research and educational purposes.

@article{spn3dpointclouds,
  author    = {Jiayun Wang and
               Rudrasis Chakraborty and
               Stella X. Yu},
  title     = {Spatial Transformer for 3D Points},
  journal   = {CoRR},
  volume    = {abs/1906.10887},
  year      = {2019},
  url       = {http://arxiv.org/abs/1906.10887},
}
You can’t perform that action at this time.