Skip to content

Pytorch unofficial implementation of PUGAN (a Point Cloud Upsampling Adversarial Network, ICCV, 2019)

Notifications You must be signed in to change notification settings

HaolinLiu97/PUGAN-pytorch

Repository files navigation

PUGAN-pytorch

Pytorch unofficial implementation of PUGAN (a Point Cloud Upsampling Adversarial Network, ICCV, 2019)

Install some packages

simply by

pip install -r requirement.txt

Install Pointnet2 module

cd pointnet2
python setup.py install

Install KNN_cuda

pip install --upgrade https://github.com/unlimblue/KNN_CUDA/releases/download/0.2/KNN_CUDA-0.2-py3-none-any.whl

dataset

We use the PU-Net dataset for training, you can refer to https://github.com/yulequan/PU-Net to download the .h5 dataset file, which can be directly used in this project.

modify some setting in the option/train_option.py

change opt['project_dir'] to where this project is located, and change opt['dataset_dir'] to where you store the dataset.
also change params['train_split'] and params['test_split'] to where you save the train/test split txt files.

training

cd train
python train.py --exp_name=the_project_name --gpu=gpu_number --use_gan --batch_size=12

About

Pytorch unofficial implementation of PUGAN (a Point Cloud Upsampling Adversarial Network, ICCV, 2019)

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published