Scalable landmark recognition using EXTENT

  • Authors:
  • Arun Qamra;Edward Y. Chang

  • Affiliations:
  • Department of Computer Science, University of California Santa Barbara, Santa Barbara, USA;VIMA Technologies, Santa Barbara, USA

  • Venue:
  • Multimedia Tools and Applications
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

We have proposed the EXTENT system for automated photograph annotation using image content and context analysis. A key component of EXTENT is a Landmark recognition system called LandMarker. In this paper, we present the architecture of LandMarker. The content of a query photograph is analyzed and compared against a database of sample landmark images, to recognize any landmarks it contains. An algorithm is presented for comparing a query image with a sample image. Context information may be used to assist landmark recognition. Also, we show how LandMarker deals with scalability to allow recognition of a large number of landmarks. We have implemented a prototype of the system, and present empirical results on a large dataset.