Pretražite po imenu i prezimenu autora, mentora, urednika, prevoditelja

Napredna pretraga

Pregled bibliografske jedinice broj: 1068608

Visual Place Recognition using Directed Acyclic Graph Association Measures and Mutual Information-based Feature Selection


Maltar, Jurica; Marković, Ivan; Petrović, Ivan
Visual Place Recognition using Directed Acyclic Graph Association Measures and Mutual Information-based Feature Selection // Robotics and autonomous systems, 132 (2020), 103598, 14 (međunarodna recenzija, članak, znanstveni)


CROSBI ID: 1068608 Za ispravke kontaktirajte CROSBI podršku putem web obrasca

Naslov
Visual Place Recognition using Directed Acyclic Graph Association Measures and Mutual Information-based Feature Selection

Autori
Maltar, Jurica ; Marković, Ivan ; Petrović, Ivan

Izvornik
Robotics and autonomous systems (0921-8890) 132 (2020); 103598, 14

Vrsta, podvrsta i kategorija rada
Radovi u časopisima, članak, znanstveni

Ključne riječi
Visual place recognition ; localization ; deep convolutional neural networks ; mutual information-based feature selection ; SeqSLAM

Sažetak
Visual localization is a challenging problem, especially over the long run, since places can exhibit significant variation due to dynamic environmental and seasonal changes. To tackle this problem, we propose a visual place recognition method based on directed acyclic graph matching and feature maps extracted from deep convolutional neural networks (DCNN). Furthermore, in order to find the best subset of DCNN feature maps with minimal redundancy, we propose to form probability distributions on image representation features and leverage the Jensen-Shannon divergence to rank features. We evaluate the proposed approach on two challenging public datasets, namely the Bonn and the Freiburg datasets, and compare it to the state-of-the-art methods. For image representations, we evaluated the following DCNN architectures: AlexNet, OverFeat, ResNet18 and ResNet50. Due to the proposed graph structure, we are able to account for any kind of correlations in image sequences, and therefore dub our approach NOSeqSLAM. Algorithms with and without feature selection were evaluated based on precision-recall curves, area under the curve score, best recall at 100% precision score and running time, with NOSeqSLAM outperforming the counterpart approaches. Furthermore, by formulating the mutual information-based feature selection specifically for visual place recognition and by selecting the feature percentile with the best score, all the algorithms, and not just NOSeqSLAM, exhibited enhanced performance with the reduced feature set.

Izvorni jezik
Engleski

Znanstvena područja
Elektrotehnika, Računarstvo, Temeljne tehničke znanosti



POVEZANOST RADA


Ustanove:
Fakultet elektrotehnike i računarstva, Zagreb,
Sveučilište u Osijeku, Odjel za matematiku

Profili:

Avatar Url Ivan Petrović (autor)

Avatar Url Ivan Marković (autor)

Avatar Url Jurica Maltar (autor)


Citiraj ovu publikaciju:

Maltar, Jurica; Marković, Ivan; Petrović, Ivan
Visual Place Recognition using Directed Acyclic Graph Association Measures and Mutual Information-based Feature Selection // Robotics and autonomous systems, 132 (2020), 103598, 14 (međunarodna recenzija, članak, znanstveni)
Maltar, J., Marković, I. & Petrović, I. (2020) Visual Place Recognition using Directed Acyclic Graph Association Measures and Mutual Information-based Feature Selection. Robotics and autonomous systems, 132, 103598, 14.
@article{article, year = {2020}, pages = {14}, chapter = {103598}, keywords = {Visual place recognition, localization, deep convolutional neural networks, mutual information-based feature selection, SeqSLAM}, journal = {Robotics and autonomous systems}, volume = {132}, issn = {0921-8890}, title = {Visual Place Recognition using Directed Acyclic Graph Association Measures and Mutual Information-based Feature Selection}, keyword = {Visual place recognition, localization, deep convolutional neural networks, mutual information-based feature selection, SeqSLAM}, chapternumber = {103598} }
@article{article, year = {2020}, pages = {14}, chapter = {103598}, keywords = {Visual place recognition, localization, deep convolutional neural networks, mutual information-based feature selection, SeqSLAM}, journal = {Robotics and autonomous systems}, volume = {132}, issn = {0921-8890}, title = {Visual Place Recognition using Directed Acyclic Graph Association Measures and Mutual Information-based Feature Selection}, keyword = {Visual place recognition, localization, deep convolutional neural networks, mutual information-based feature selection, SeqSLAM}, chapternumber = {103598} }

Časopis indeksira:


  • Current Contents Connect (CCC)
  • Web of Science Core Collection (WoSCC)
    • Science Citation Index Expanded (SCI-EXP)
    • SCI-EXP, SSCI i/ili A&HCI
  • Scopus





Contrast
Increase Font
Decrease Font
Dyslexic Font