thumbnail

Reconstructing and Exploring Massive Detailed Cityscapes

Enrico Gobbetti, Fabio Marton, Marco Di Benedetto, Fabio Ganovelli, Matthias Buehler, Simon Schubiger, Matthias Specht, Chris Engels, and Luc Van Gool

October 2011

Abstract

We present a state-of-the-art system for obtaining and exploring large scale thr ee-dimensional models of urban landscapes. A multimodal approach to reconstructi on fuses cadastral information, laser range data, and oblique imagery into build ing models, which are then refined by applying procedural rules for replacing te xtures with 3D elements, such as windows and doors, therefore enhancing the mode l quality and adding semantics to the model. For city scale exploration, these d etailed models are uploaded to a web-based service, which automatically construc ts an approximate scalable multiresolution representation. This representation c an be interactively transmitted and visualized over the net to clients ranging f rom graphics PCs to web-enabled portable devices. The approach's characteristics and performance are illustrated using real-world city-scale data.

Reference and download information

Enrico Gobbetti, Fabio Marton, Marco Di Benedetto, Fabio Ganovelli, Matthias Buehler, Simon Schubiger, Matthias Specht, Chris Engels, and Luc Van Gool. Reconstructing and Exploring Massive Detailed Cityscapes. In The 12th International Symposium on Virtual Reality, Archaeology and Cultural Heritage. Pages 1-8, October 2011.

Related multimedia productions

Bibtex citation record

@InProceedings{Gobbetti:2011:REM,
    author = {Enrico Gobbetti and Fabio Marton and Marco {Di Benedetto} and Fabio Ganovelli and Matthias Buehler and Simon Schubiger and Matthias Specht and Chris Engels and Luc Van Gool},
    title = {Reconstructing and Exploring Massive Detailed Cityscapes},
    booktitle = {The 12th International Symposium on Virtual Reality, Archaeology and Cultural Heritage},
    pages = {1-8},
    month = {October},
    year = {2011},
    isbn = {978-3-905674-34-7},
    issn = {1811-864X},
    abstract = { We present a state-of-the-art system for obtaining and exploring large scale thr ee-dimensional models of urban landscapes. A multimodal approach to reconstructi on fuses cadastral information, laser range data, and oblique imagery into build ing models, which are then refined by applying procedural rules for replacing te xtures with 3D elements, such as windows and doors, therefore enhancing the mode l quality and adding semantics to the model. For city scale exploration, these d etailed models are uploaded to a web-based service, which automatically construc ts an approximate scalable multiresolution representation. This representation c an be interactively transmitted and visualized over the net to clients ranging f rom graphics PCs to web-enabled portable devices. The approach's characteristics and performance are illustrated using real-world city-scale data. },
    url = {http://vic.crs4.it/vic/cgi-bin/bib-page.cgi?id='Gobbetti:2011:REM'},
}