Skip to content

hrzhou2/seedformer

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

SeedFormer: Patch Seeds based Point Cloud Completion with Upsample Transformer

example

This repository contains PyTorch implementation for SeedFormer: Patch Seeds based Point Cloud Completion with Upsample Transformer (ECCV 2022).

SeedFormer presents a novel method for Point Cloud Completion. In this work, we introduce a new shape representation, namely Patch Seeds, which not only captures general structures from partial inputs but also preserves regional information of local patterns. Moreover, we devise a novel Upsample Transformer by extending the transformer structure into basic operations of point generators, which explicitly incorporates spatial and semantic relationships in the local neighborhood.

If you find our work useful in your research, please cite:

@article{zhou2022seedformer,
  title={SeedFormer: Patch Seeds based Point Cloud Completion with Upsample Transformer},
  author={Zhou, Haoran and Cao, Yun and Chu, Wenqing and Zhu, Junwei and Lu, Tong and Tai, Ying and Wang, Chengjie},
  journal={arXiv preprint arXiv:2207.10315},
  year={2022}
}

🔥Updates

  • 2022-07-05: Initial Update.
  • 2022-08-27: Update pretrained models and results.

Installation

The code has been tested on one configuration:

  • python == 3.6.8
  • PyTorch == 1.8.1
  • CUDA == 10.2
  • numpy
  • open3d
pip install -r requirements.txt

Compile the C++ extension modules:

sh install.sh

Datasets

The details of used datasets can be found in DATASET.md (we thank the authors of PoinTr).

Pretrained Models and Results

We provide our generated complete point clouds on pcn testset here.

SeedFormer pretrained models are very light generator models 12.8 MB.

Usage

Training on PCN dataset

First, you should specify your dataset directories in train_pcn.py:

__C.DATASETS.SHAPENET.PARTIAL_POINTS_PATH        = '<*PATH-TO-YOUR-DATASET*>/PCN/%s/partial/%s/%s/%02d.pcd'
__C.DATASETS.SHAPENET.COMPLETE_POINTS_PATH       = '<*PATH-TO-YOUR-DATASET*>/PCN/%s/complete/%s/%s.pcd'

SeedFormer takes two V100 gpus with a batch size of 48. To train SeedFormer on PCN dataset, simply run:

python3 train_pcn.py

Testing on PCN dataset

To test a pretrained model, run:

python3 train_pcn.py --test

Or you can give the model directory name to test one particular model:

python3 train_pcn.py --test --pretrained train_pcn_Log_2022_XX_XX_XX_XX_XX

Save generated complete point clouds as well as gt and partial clouds in testing:

python3 train_pcn.py --test --output 1

Using ShapeNet-55/34

To use ShapeNet55 dataset, change the data directoriy in train_shapenet55.py:

__C.DATASETS.SHAPENET55.COMPLETE_POINTS_PATH     = '<*PATH-TO-YOUR-DATASET*>/ShapeNet55/shapenet_pc/%s'

Then, run:

python3 train_shapenet55.py

In order to switch to ShapeNet34, you can change the data file in train_shapenet55.py:

__C.DATASETS.SHAPENET55.CATEGORY_FILE_PATH       = './datasets/ShapeNet55-34/ShapeNet-34/'

The testing process is very similar to that on PCN:

python3 train_shapenet55.py --test

Acknowledgement

Some parts of the code are borrowed from GRNet and SnowflakeNet. We thank the authors for their excellent work.

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published