Camera Elevation Estimation from a Single Mountain Landscape Photograph

Martin Čadík, Jan Vašíček, Michal Hradiš, Filip Radenović and Ondřej Chum

Abstract

This work addresses the problem of camera elevation estimation from a single photograph in an outdoor environment. We introduce a new benchmark dataset of one-hundred thousand images with annotated camera elevation called Alps100K. We propose and experimentally evaluate two automatic data-driven approaches to camera elevation estimation: one based on convolutional neural networks, the other on local features. To compare the proposed methods to human performance, an experiment with 100 subjects is conducted. The experimental results show that both proposed approaches outperform humans and that the best result is achieved by their combination.

Session

Poster 1

Files

PDF iconExtended Abstract (PDF, 1616K)
PDF iconPaper (PDF, 2M)

DOI

10.5244/C.29.30
https://dx.doi.org/10.5244/C.29.30

Citation

Martin Čadík, Jan Vašíček, Michal Hradiš, Filip Radenović and Ondřej Chum. Camera Elevation Estimation from a Single Mountain Landscape Photograph. In Xianghua Xie, Mark W. Jones, and Gary K. L. Tam, editors, Proceedings of the British Machine Vision Conference (BMVC), pages 30.1-30.12. BMVA Press, September 2015.

Bibtex

@inproceedings{BMVC2015_30,
	title={Camera Elevation Estimation from a Single Mountain Landscape Photograph},
	author={Martin Čadík and Jan Vašíček and Michal Hradiš and Filip Radenović and Ondřej Chum},
	year={2015},
	month={September},
	pages={30.1-30.12},
	articleno={30},
	numpages={12},
	booktitle={Proceedings of the British Machine Vision Conference (BMVC)},
	publisher={BMVA Press},
	editor={Xianghua Xie, Mark W. Jones, and Gary K. L. Tam},
	doi={10.5244/C.29.30},
	isbn={1-901725-53-7},
	url={https://dx.doi.org/10.5244/C.29.30}
}