City-Scale Location Recognition

Grant Schindler; Matthew Brown; Richard Szeliski

doi:10.1109/CVPR.2007.383150

City-Scale Location Recognition

Grant Schindler, Matthew Brown, Richard Szeliski

Department of Computer Science

Research output: Contribution to conference › Paper › peer-review

540 Citations (SciVal)

Abstract

We look at the problem of location recognition in a large image dataset using a vocabulary tree. This entails finding the location of a query image in a large dataset containing 3times104 streetside images of a city. We investigate how the traditional invariant feature matching approach falls down as the size of the database grows. In particular we show that by carefully selecting the vocabulary using the most informative features, retrieval performance is significantly improved, allowing us to increase the number of database images by a factor of 10. We also introduce a generalization of the traditional vocabulary tree search algorithm which improves performance by effectively increasing the branching factor of a fixed vocabulary tree.

Original language	English
DOIs	https://doi.org/10.1109/CVPR.2007.383150
Publication status	Published - Jun 2007
Event	CVPR '07: IEEE Conference on Computer Vision and Pattern Recognition, 2007 - Minneapolis Duration: 17 Jun 2007 → 22 Jun 2007

Conference

Conference	CVPR '07: IEEE Conference on Computer Vision and Pattern Recognition, 2007
City	Minneapolis
Period	17/06/07 → 22/06/07

Access to Document

10.1109/CVPR.2007.383150

Cite this

@conference{5b882aaa5b224ee69d79a58d2780473c,

title = "City-Scale Location Recognition",

abstract = "We look at the problem of location recognition in a large image dataset using a vocabulary tree. This entails finding the location of a query image in a large dataset containing 3times104 streetside images of a city. We investigate how the traditional invariant feature matching approach falls down as the size of the database grows. In particular we show that by carefully selecting the vocabulary using the most informative features, retrieval performance is significantly improved, allowing us to increase the number of database images by a factor of 10. We also introduce a generalization of the traditional vocabulary tree search algorithm which improves performance by effectively increasing the branching factor of a fixed vocabulary tree.",

author = "Grant Schindler and Matthew Brown and Richard Szeliski",

year = "2007",

month = jun,

doi = "10.1109/CVPR.2007.383150",

language = "English",

note = "CVPR '07: IEEE Conference on Computer Vision and Pattern Recognition, 2007 ; Conference date: 17-06-2007 Through 22-06-2007",

}

TY - CONF

T1 - City-Scale Location Recognition

AU - Schindler, Grant

AU - Brown, Matthew

AU - Szeliski, Richard

PY - 2007/6

Y1 - 2007/6

N2 - We look at the problem of location recognition in a large image dataset using a vocabulary tree. This entails finding the location of a query image in a large dataset containing 3times104 streetside images of a city. We investigate how the traditional invariant feature matching approach falls down as the size of the database grows. In particular we show that by carefully selecting the vocabulary using the most informative features, retrieval performance is significantly improved, allowing us to increase the number of database images by a factor of 10. We also introduce a generalization of the traditional vocabulary tree search algorithm which improves performance by effectively increasing the branching factor of a fixed vocabulary tree.

AB - We look at the problem of location recognition in a large image dataset using a vocabulary tree. This entails finding the location of a query image in a large dataset containing 3times104 streetside images of a city. We investigate how the traditional invariant feature matching approach falls down as the size of the database grows. In particular we show that by carefully selecting the vocabulary using the most informative features, retrieval performance is significantly improved, allowing us to increase the number of database images by a factor of 10. We also introduce a generalization of the traditional vocabulary tree search algorithm which improves performance by effectively increasing the branching factor of a fixed vocabulary tree.

UR - http://dx.doi.org/10.1109/CVPR.2007.383150

U2 - 10.1109/CVPR.2007.383150

DO - 10.1109/CVPR.2007.383150

M3 - Paper

T2 - CVPR '07: IEEE Conference on Computer Vision and Pattern Recognition, 2007

Y2 - 17 June 2007 through 22 June 2007

ER -

City-Scale Location Recognition

Abstract

Conference

Access to Document

Other files and links

Fingerprint

Cite this