Visual Place Recognition in Changing Environments with Sequence Representations on the Distance-Space Domain
Navigating in a perpetually changing world can provide the basis for numerous challenging autonomous robotic applications. With a view to long-term autonomy, visual place recognition (vPR) systems should be able to robustly operate under extreme appearance changes in their environment. Typically, th...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-05-01
|
| Series: | Machines |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2075-1702/11/5/558 |
| _version_ | 1797599395758211072 |
|---|---|
| author | Ioannis Tsampikos Papapetros Ioannis Kansizoglou Loukas Bampis Antonios Gasteratos |
| author_facet | Ioannis Tsampikos Papapetros Ioannis Kansizoglou Loukas Bampis Antonios Gasteratos |
| author_sort | Ioannis Tsampikos Papapetros |
| collection | DOAJ |
| description | Navigating in a perpetually changing world can provide the basis for numerous challenging autonomous robotic applications. With a view to long-term autonomy, visual place recognition (vPR) systems should be able to robustly operate under extreme appearance changes in their environment. Typically, the utilized data representations are heavily influenced by those changes, negatively affecting the vPR performance. In this article, we propose a sequence-based technique that decouples such changes from the similarity estimation procedure. This is achieved by remapping the sequential representation data into the distance-space domain, i.e., a domain in which we solely consider the distances between image instances, and subsequently normalize them. In such a way, perturbations related to different environmental conditions and embedded into the original representation vectors are avoided, therefore the scene recognition efficacy is enhanced. We evaluate our framework under multiple different instances, with results indicating a significant performance improvement over other approaches. |
| first_indexed | 2024-03-11T03:33:50Z |
| format | Article |
| id | doaj.art-c70f058d3dd4493c8c138f96d367318f |
| institution | Directory Open Access Journal |
| issn | 2075-1702 |
| language | English |
| last_indexed | 2024-03-11T03:33:50Z |
| publishDate | 2023-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Machines |
| spelling | doaj.art-c70f058d3dd4493c8c138f96d367318f2023-11-18T02:11:52ZengMDPI AGMachines2075-17022023-05-0111555810.3390/machines11050558Visual Place Recognition in Changing Environments with Sequence Representations on the Distance-Space DomainIoannis Tsampikos Papapetros0Ioannis Kansizoglou1Loukas Bampis2Antonios Gasteratos3Department of Production and Management Engineering, Democritus University of Thrace, Vas. Sophias 12, GR-671 32 Xanthi, GreeceDepartment of Production and Management Engineering, Democritus University of Thrace, Vas. Sophias 12, GR-671 32 Xanthi, GreeceDepartment of Electrical and Computer Engineering, Democritus University of Thrace, Building B, Kimmeria Campus, GR-671 32 Xanthi, GreeceDepartment of Production and Management Engineering, Democritus University of Thrace, Vas. Sophias 12, GR-671 32 Xanthi, GreeceNavigating in a perpetually changing world can provide the basis for numerous challenging autonomous robotic applications. With a view to long-term autonomy, visual place recognition (vPR) systems should be able to robustly operate under extreme appearance changes in their environment. Typically, the utilized data representations are heavily influenced by those changes, negatively affecting the vPR performance. In this article, we propose a sequence-based technique that decouples such changes from the similarity estimation procedure. This is achieved by remapping the sequential representation data into the distance-space domain, i.e., a domain in which we solely consider the distances between image instances, and subsequently normalize them. In such a way, perturbations related to different environmental conditions and embedded into the original representation vectors are avoided, therefore the scene recognition efficacy is enhanced. We evaluate our framework under multiple different instances, with results indicating a significant performance improvement over other approaches.https://www.mdpi.com/2075-1702/11/5/558visual place recognitionchanging environmentssequence matchinglocalizationnavigation |
| spellingShingle | Ioannis Tsampikos Papapetros Ioannis Kansizoglou Loukas Bampis Antonios Gasteratos Visual Place Recognition in Changing Environments with Sequence Representations on the Distance-Space Domain Machines visual place recognition changing environments sequence matching localization navigation |
| title | Visual Place Recognition in Changing Environments with Sequence Representations on the Distance-Space Domain |
| title_full | Visual Place Recognition in Changing Environments with Sequence Representations on the Distance-Space Domain |
| title_fullStr | Visual Place Recognition in Changing Environments with Sequence Representations on the Distance-Space Domain |
| title_full_unstemmed | Visual Place Recognition in Changing Environments with Sequence Representations on the Distance-Space Domain |
| title_short | Visual Place Recognition in Changing Environments with Sequence Representations on the Distance-Space Domain |
| title_sort | visual place recognition in changing environments with sequence representations on the distance space domain |
| topic | visual place recognition changing environments sequence matching localization navigation |
| url | https://www.mdpi.com/2075-1702/11/5/558 |
| work_keys_str_mv | AT ioannistsampikospapapetros visualplacerecognitioninchangingenvironmentswithsequencerepresentationsonthedistancespacedomain AT ioanniskansizoglou visualplacerecognitioninchangingenvironmentswithsequencerepresentationsonthedistancespacedomain AT loukasbampis visualplacerecognitioninchangingenvironmentswithsequencerepresentationsonthedistancespacedomain AT antoniosgasteratos visualplacerecognitioninchangingenvironmentswithsequencerepresentationsonthedistancespacedomain |