Skip to main navigation Skip to search Skip to main content

Visual Place Recognition Using Landmark Distribution Descriptors

  • University of Bristol

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

6 Citations (Scopus)

Abstract

Recent work by Sünderhauf et al. [1] demonstrated improved visual place recognition using proposal regions coupled with features from convolutional neural networks (CNN) to match landmarks between views. In this work we extend the approach by introducing descriptors built from landmark features which also encode the spatial distribution of the landmarks within a view. Matching descriptors then enforces consistency of the relative positions of landmarks between views. This has a significant impact on performance. For example, in experiments on 10 image-pair datasets, each consisting of 200 urban locations with significant differences in viewing positions and conditions, we recorded average precision of around 70% (at 100% recall), compared with 58% obtained using whole image CNN features and 50% for the method in [1].

Original languageEnglish
Title of host publicationComputer Vision – ACCV 2016 - 13th Asian Conference on Computer Vision, Revised Selected Papers, Part 4
EditorsShang-Hong Lai, Ko Nishino, Vincent Lepetit, Yoichi Sato
PublisherSpringer Science and Business Media Deutschland GmbH
Pages487-502
Number of pages16
ISBN (Print)9783319541891
DOIs
Publication statusPublished - 2017
Externally publishedYes
Event13th Asian Conference on Computer Vision, ACCV 2016 - Taipei, Taiwan, Province of China
Duration: 20 Nov 201624 Nov 2016

Publication series

NameLecture Notes in Computer Science
Volume10114 LNIP
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference13th Asian Conference on Computer Vision, ACCV 2016
Country/TerritoryTaiwan, Province of China
City Taipei
Period20/11/1624/11/16

Fingerprint

Dive into the research topics of 'Visual Place Recognition Using Landmark Distribution Descriptors'. Together they form a unique fingerprint.

Cite this