PU-GAN: a Point Cloud Upsampling Adversarial Network

by Ruihui Li, Xianzhi Li, Chi-Wing Fu, Daniel Cohen-Or, Pheng-Ann Heng.

Introduction

This repository is for our ICCV 2019 paper 'PU-GAN: a Point Cloud Upsampling Adversarial Network'. The code is modified from 3PU and PU-Net.

Docker

A Dockerfile is provided to help you relief the pain of configurate training environment.

See the instructions in here.

Installation

This repository is based on Tensorflow and the TF operators from PointNet++. Therefore, you need to install tensorflow and compile the TF operators.

For installing tensorflow, please follow the official instructions in here. The code is tested under TF1.11 (higher version should also work) and Python 3.6 on Ubuntu 16.04.

For compiling TF operators, please check tf_xxx_compile.sh under each op subfolder in code/tf_ops folder. Note that you need to update nvcc, python and tensoflow include library if necessary.

Note

When running the code, if you have undefined symbol: _ZTIN10tensorflow8OpKernelE error, you need to compile the TF operators. If you have already added the -I$TF_INC/external/nsync/public -L$TF_LIB -ltensorflow_framework but still have cannot find -ltensorflow_framework error. Please use 'locate tensorflow_framework ' to locate the tensorflow_framework library and make sure this path is in $TF_LIB.

Usage

Clone the repository:

https://github.com/liruihui/PU-GAN.git
cd PU-GAN

Compile the TF operators Follow the above information to compile the TF operators.
Train the model: First, you need to download the training patches in HDF5 format from GoogleDrive and put it in folder data/train. Then run:
```
cd code
python pu_gan.py --phase train
```
Evaluate the model: First, you need to download the pretrained model from GoogleDrive, extract it and put it in folder 'model'. Then run:
```
cd code
python pu_gan.py --phase test
```
You will see the input and output results in the folder data/test/output.
The training and testing mesh files can be downloaded from GoogleDrive.

Evaluation code

We provide the code to calculate the uniform metric in the evaluation code folder. In order to use it, you need to install the CGAL library. Please refer this link and PU-Net to install this library. Then:

cd evaluation_code
cmake .
make
./evaluation Icosahedron.off Icosahedron.xyz

The second argument is the mesh, and the third one is the predicted points.

Citation

If PU-GAN is useful for your research, please consider citing:

@inproceedings{li2019pugan,
     title={PU-GAN: a Point Cloud Upsampling Adversarial Network},
     author={Li, Ruihui and Li, Xianzhi and Fu, Chi-Wing and Cohen-Or, Daniel and Heng, Pheng-Ann},
     booktitle = {{IEEE} International Conference on Computer Vision ({ICCV})},
     year = {2019}
 }

Questions

Please contact 'lirh@cse.cuhk.edu.hk'