Online reconstruction of textured triangle meshes from aerial images

Authors:
Tom Vierjahn;Jan Roters;Manuel Moser;Klaus Hinrichs;Sina Mostafawy
Affiliations:
[rmh] new media GmbH, Cologne, Germany and University of Muenster, Germany and FH Duesseldorf University of Applied Sciences, Germany;University of Muenster, Germany;[rmh] new media GmbH, Cologne, Germany;University of Muenster, Germany;[rmh] new media GmbH, Cologne, Germany and FH Duesseldorf University of Applied Sciences, Germany
Venue:
UDMV '13 Proceedings of the Eurographics Workshop on Urban Data Modelling and Visualisation
Year:
2013

Citing 2
Cited 0

Growing self-reconstruction maps

IEEE Transactions on Neural Networks
Building Rome on a cloudless day

ECCV'10 Proceedings of the 11th European conference on Computer vision: Part IV

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper we present and evaluate a new online reconstruction algorithm to create a textured triangle mesh from a set of aerial images via an unorganized point cloud. Both the point cloud and the mesh are iteratively refined while allowing new aerial images to be added at any time during reconstruction. Texture coordinates are learnt to instantly visualize an initially rough approximation that gets refined as more data becomes available. The new algorithm improves upon other systems that require the complete data to be acquired beforehand, and that apply offline, non-iterative reconstruction and processing. Thus, our algorithm is perfectly suited for time-critical applications, e. g., strategical visualization platforms for disaster and emergency response.