Automatic photo-to-terrain alignment for the annotation of mountain pictures

Abstract

The researchers of the Max-Planck-Institut für Informatik, in collaboration with Télécom ParisTech, finalized their work on automatic alignment that will be presented on the 21st of June at the prestigious CVPR conference. The goal of this method is to register arbitrary mountain pictures and movies from the internet into a Google-Earth-like 3D Model. Such a technique could be used by travelers to transform their pictures into a virtual and realistic 3D experience (like Google Streetview). In particular, the system could also be used to automatically highlight elements in the image, such as the travel path taken, names of mountains, or other landmarks. The work could further be used to augment the realism and level of detail of applications such as Google Earth by transferring the information from photo and movies directly onto the 3D model.

DOI: 10.1109/CVPR.2011.5995727
View Slides

Extracted Key Phrases

1 Figure or Table

010202011201220132014201520162017
Citations per Year

61 Citations

Semantic Scholar estimates that this publication has 61 citations based on the available data.

See our FAQ for additional information.

Cite this paper

@inproceedings{Baboud2011AutomaticPA, title={Automatic photo-to-terrain alignment for the annotation of mountain pictures}, author={Lionel Baboud and Martin Cad{\'i}k and Elmar Eisemann and Hans-Peter Seidel}, booktitle={CVPR}, year={2011} }