Building Rome on a Cloudless Day
Abstract
This paper introduces an approach for dense 3D reconstruction from unregistered Internet-scale photo collections with about 3 million images within the span of a day on a single PC (“cloudless”). Our method advances image clustering, stereo, stereo fusion and structure from motion to achieve high computational performance. We leverage geometric and appearance constraints to obtain a highly parallel implementation on modern graphics processors and multi-core architectures. This leads to two orders of magnitude higher performance on an order of magnitude larger dataset than competing state-of-the-art approaches.
Cite
Text
Frahm et al. "Building Rome on a Cloudless Day." European Conference on Computer Vision, 2010. doi:10.1007/978-3-642-15561-1_27Markdown
[Frahm et al. "Building Rome on a Cloudless Day." European Conference on Computer Vision, 2010.](https://mlanthology.org/eccv/2010/frahm2010eccv-building/) doi:10.1007/978-3-642-15561-1_27BibTeX
@inproceedings{frahm2010eccv-building,
title = {{Building Rome on a Cloudless Day}},
author = {Frahm, Jan-Michael and Georgel, Pierre Fite and Gallup, David and Johnson, Tim and Raguram, Rahul and Wu, Changchang and Jen, Yi-Hung and Dunn, Enrique and Clipp, Brian and Lazebnik, Svetlana},
booktitle = {European Conference on Computer Vision},
year = {2010},
pages = {368-381},
doi = {10.1007/978-3-642-15561-1_27},
url = {https://mlanthology.org/eccv/2010/frahm2010eccv-building/}
}