SfM Flow is a comprehensive toolset for the evaluation of 3D reconstruction pipelines. It provides tools for the creation of datasets composed of synthetic images, the execution of 3D reconstruction Structure from Motion (SfM) pipelines, and the evaluation of the obtained results. SfM Flow integrates all these features in Blender.
Read more in the paper SfM Flow: A comprehensive toolset for the evaluation of 3D reconstruction pipelines by Davide Marelli, Simone Bianco and Gianluigi Ciocca.
- Blender 2.80.75 or later (Blender 2.93 LTS recommended) - tested on Ubuntu 18.04, Windows 10 and macOS Mojave
- ExifTool v10 or later
- A Structure from Motion 3D reconstruction pipeline, see pipeline support
See the Getting started wiki page.
Read the wiki for further details.
If you use this software, please cite these papers:
@article{marelli2022sfmflow,
author = {Marelli, Davide and Bianco, Simone and Ciocca Gianluigi},
year = {2022},
pages = {100931},
title = {SfM Flow: A comprehensive toolset for the evaluation of 3D reconstruction pipelines},
volume = {17},
journal = {SoftwareX},
doi = {10.1016/j.softx.2021.100931},
issn = {2352-7110},
projectref = {http://www.ivl.disco.unimib.it/activities/evaluating-the-performance-of-structure-from-motion-pipelines/}
}
@article{bianco-sfm2018,
author = {Bianco, Simone and Ciocca, Gianluigi and Marelli, Davide},
year = {2018},
title = {Evaluating the Performance of Structure from Motion Pipelines},
volume = {4},
number = {8},
journal = {Journal of Imaging},
doi = {10.3390/jimaging4080098},
projectref = {http://www.ivl.disco.unimib.it/activities/evaluating-the-performance-of-structure-from-motion-pipelines/}
}
@inproceedings{bianco2018blender-plugin,
author = {Marelli, Davide and Bianco, Simone and Celona, Luigi and Ciocca, Gianluigi},
year = {2018},
pages = {1-5},
title = {A Blender plug-in for comparing Structure from Motion pipelines},
organization = {IEEE},
booktitle = {2018 IEEE 8th International Conference on Consumer Electronics - Berlin (ICCE-Berlin)},
doi = {10.1109/ICCE-Berlin.2018.8576196},
issn = {2166-6822}
}