Learning and Calibrating Per Location Classifiers for Visual Place Recognition (Open Access)

Abstract

The aim of this work is to localize a query photograph by finding other images depicting the same place in a large geotagged image database. This is a challenging task due to changes in viewpoint, imaging conditions and the large size of the image database. The contribution of this work is two-fold. First, we cast the place recognition problem as a classification task and use the available geotags to train a classifier for each location in the database in a similar manner to per-exemplar SVMs in object recognition. Second, as only few positive training examples are available for each location, we propose a new approach to calibrate all the per-location SVM classifiers using only the negative examples. The calibration we propose relies on a significance measure essentially equivalent to the p-values classically used in statistical hypothesis testing. Experiments are performed on a database of 25,000 geotagged street view images of Pittsburgh and demonstrate improved place recognition accuracy of the proposed approach over the previous work.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jun 23, 2013
Accession Number
AD1039705

Entities

People

  • Guillaume Obozinski
  • Josef Sivic
  • Petr Gronat
  • Tomas Pajdla

Tags

Communities of Interest

  • C4I
  • Energy and Power Technologies

DTIC Thesaurus Topics

  • Air Force Research Laboratories
  • Calibration
  • Classification
  • Computer Vision
  • Databases
  • Electrical Engineering
  • Images
  • Information Science
  • Learning
  • Machine Learning
  • Machine Perception
  • Object Recognition
  • Pattern Recognition
  • Probability
  • Random Variables
  • Recognition
  • Supervised Machine Learning

Fields of Study

  • Computer science

Readers

  • Computer Vision.
  • Regression Analysis.