Learning and Calibrating Per-Location Classifiers for Visual Place Recognition

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F16%3A00304267" target="_blank" >RIV/68407700:21230/16:00304267 - isvavai.cz</a>

Výsledek na webu

<a href="http://link.springer.com/article/10.1007/s11263-015-0878-x" target="_blank" >http://link.springer.com/article/10.1007/s11263-015-0878-x</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1007/s11263-015-0878-x" target="_blank" >10.1007/s11263-015-0878-x</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Learning and Calibrating Per-Location Classifiers for Visual Place Recognition

Popis výsledku v původním jazyce

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 one or a few positive training examples are available for each location, we propose two methods to calibrate all the per-location SVM classifiers without the need for additional positive training data. The first method relies on p-values from statistical hypothesis testing and uses only the available negative training data. The second method performs an affine calibration by appropriately normalizing the learnt classifier hyperplane and does not need any additional labelled training data. We test the proposed place recognitionmethod with the bag-of-visual-words and Fisher vector image representations suitable for large scale indexing. Experiments are performed on three datasets: 25,000 and 55,000 geotagged street view images of Pittsburgh, and the 24/7 Tokyo benchmark containing 76,000 images with varying illumination conditions. The results show improved place recognition accuracy of the learnt image representation over direct matching of raw image descriptors.

Název v anglickém jazyce

Learning and Calibrating Per-Location Classifiers for Visual Place Recognition

Popis výsledku anglicky

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 one or a few positive training examples are available for each location, we propose two methods to calibrate all the per-location SVM classifiers without the need for additional positive training data. The first method relies on p-values from statistical hypothesis testing and uses only the available negative training data. The second method performs an affine calibration by appropriately normalizing the learnt classifier hyperplane and does not need any additional labelled training data. We test the proposed place recognitionmethod with the bag-of-visual-words and Fisher vector image representations suitable for large scale indexing. Experiments are performed on three datasets: 25,000 and 55,000 geotagged street view images of Pittsburgh, and the 24/7 Tokyo benchmark containing 76,000 images with varying illumination conditions. The results show improved place recognition accuracy of the learnt image representation over direct matching of raw image descriptors.

Druh

J_x - Nezařazeno - Článek v odborném periodiku (Jimp, Jsc a Jost)

CEP obor

JD - Využití počítačů, robotika a její aplikace

OECD FORD obor

—

Projekt

<a href="/cs/project/7E13015" target="_blank" >7E13015: Planetary Robotics Data Exploitation</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2016

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název periodika

International Journal of Computer Vision

ISSN

0920-5691

e-ISSN

—

Svazek periodika

118

Číslo periodika v rámci svazku

3

Stát vydavatele periodika

NL - Nizozemsko

Počet stran výsledku

18

Strana od-do

319-336

Kód UT WoS článku

000378789100003

EID výsledku v databázi Scopus

—