Home

Awesome

Global Transport for Fluid Reconstruction with Learned Self-Supervision

This repository is the official implementation of Global Transport for Fluid Reconstruction with Learned Self-Supervision (Project Website, arXiv).

Overview Image
Abstract
We propose a novel method to reconstruct volumetric flows from sparse views via a global transport formulation. Instead of obtaining the space-time function of the observations, we reconstruct its motion based on a single initial state. In addition we introduce a learned self-supervision that constrains observations from unseen angles. These visual constraints are coupled via the transport constraints and a differentiable rendering step to arrive at a robust end-to-end reconstruction algorithm. This makes the reconstruction of highly realistic flow motions possible, even from only a single input view. We show with a variety of synthetic and real flows that the proposed global reconstruction of the transport process yields an improved reconstruction of the fluid motion.

Requirements

Installation

Reconstruction

To reconstruct the final results of the paper, run this command:

You can disable console output with the -c option.

Evaluation

Basic evaluation and rendering of the reconstruction is done automatically at the end of the reconstruction procedure.
To further evaluate, take a look here:

python eval_runs.py -h

Results

Sample Results
Left: multi-view, right: single-view.

Citation

@InProceedings{Franz_2021_CVPR,
    author    = {Franz, Erik and Solenthaler, Barbara and Thuerey, Nils},
    title     = {Global Transport for Fluid Reconstruction With Learned Self-Supervision},
    booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
    month     = {June},
    year      = {2021},
    pages     = {1632-1642}
}

Acknowledgements

This work was supported by the Siemens/IAS Fellowship Digital Twin, and the ERC Consolidator Grant SpaTe (CoG-2019-863850).