English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Visual place recognition (VPR) enables autonomous systems to localizethemselves within an environment using image information. While VPR techniquesbuilt upon a Convolutional Neural Network (CNN) backbone dominatestate-of-the-art VPR performance, their high computational requirements makethem unsuitable for platforms equipped with low-end hardware. Recently, alightweight VPR system based on multiple bio-inspired classifiers, dubbedDrosoNets, has been proposed, achieving great computational efficiency at thecost of reduced absolute place retrieval performance. In this work, we proposea novel multi-DrosoNet localization system, dubbed RegionDrosoNet, withsignificantly improved VPR performance, while preserving a low-computationalprofile. Our approach relies on specializing distinct groups of DrosoNets ondifferently sliced partitions of the original image, increasing extrinsic modeldifferentiation. Furthermore, we introduce a novel voting module to combine theoutputs of all DrosoNets into the final place prediction which considersmultiple top refence candidates from each DrosoNet. RegionDrosoNet outperformsother lightweight VPR techniques when dealing with both appearance changes andviewpoint variations. Moreover, it competes with computationally expensivemethods on some benchmark datasets at a small fraction of their onlineinference time.

Aggregating Multiple Bio-Inspired Image Region Classifiers For Effective And Lightweight Visual Place Recognition