GotFlow3D

This repository contains the source code GotFlow3D v1.0 for our paper: "GotFlow3D: Recurrent Graph Optimal Transport for Learning 3D Flow Motion in Particle Tracking".

Requirements

• The code is tested on Ubuntu 20.04.
• CUDA 11.3
• Python 3.8
• pytorch 1.10
• numpy
• scipy
• tqdm
• torch-scatter
• tensorboard
• imageio
• setuptools 59.5.0

Installation

conda create -n gotflow3d python=3.8
conda activate gotflow3d
conda install pytorch==1.10.1 torchvision==0.11.2 torchaudio==0.10.1 cudatoolkit=11.3 -c pytorch -c conda-forge
conda install tqdm tensorboard scipy imageio
pip install torch-scatter==2.0.9 -f https://pytorch-geometric.com/whl/torch-1.10.1+cu113.html

It can be installed within half an hour, which mainly depends on the internet speed.

Required Data

To evaluate/train GotFlow3D, you will need to download the required datasets FluidFlow3D-family. We also provide a small dataset in the data folder to demo the software/code.

Usage

Train

python train.py --exp_path=test --num_epochs=30 --iters=8 --root=./

where exp_path is the experiment folder name, num_epochs is the number of epochs in training and iters denotes the number of iterations.

Test

python test.py --exp_path=test --iters=8 --root=./ --weights=./experiments/weights_GotFlow3D/checkpoints/best_checkpoint.params

It takes about 0.1s to estimate the flow of one sample containing about 2000 particles on the NVIDIA RTX 3090 GPU. The expected output of the network is the the dense flow field.

Citation

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

@article{
  title={{GotFlow3D: Recurrent Graph Optimal Transport for Learning 3D Flow Motion in Particle Tracking}},
  author={Liang, Jiaming and Cai, Shengze and Xu, Chao},
}