Improved Visual Localization via Graph Filtering
Vision-based localization is the problem of inferring the pose of the camera given a single image. One commonly used approach relies on image retrieval where the query input is compared against a database of localized support examples and its pose is inferred with the help of the retrieved items. Th...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-01-01
|
| Series: | Journal of Imaging |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2313-433X/7/2/20 |
| _version_ | 1797405861653512192 |
|---|---|
| author | Carlos Lassance Yasir Latif Ravi Garg Vincent Gripon Ian Reid |
| author_facet | Carlos Lassance Yasir Latif Ravi Garg Vincent Gripon Ian Reid |
| author_sort | Carlos Lassance |
| collection | DOAJ |
| description | Vision-based localization is the problem of inferring the pose of the camera given a single image. One commonly used approach relies on image retrieval where the query input is compared against a database of localized support examples and its pose is inferred with the help of the retrieved items. This assumes that images taken from the same places consist of the same landmarks and thus would have similar feature representations. These representations can learn to be robust to different variations in capture conditions like time of the day or weather. In this work, we introduce a framework which aims at enhancing the performance of such retrieval-based localization methods. It consists in taking into account additional information available, such as GPS coordinates or temporal proximity in the acquisition of the images. More precisely, our method consists in constructing a graph based on this additional information that is later used to improve reliability of the retrieval process by filtering the feature representations of support and/or query images. We show that the proposed method is able to significantly improve the localization accuracy on two large scale datasets, as well as the mean average precision in classical image retrieval scenarios. |
| first_indexed | 2024-03-09T03:17:21Z |
| format | Article |
| id | doaj.art-d5188fd600da4d0ba25183fad6515e2e |
| institution | Directory Open Access Journal |
| issn | 2313-433X |
| language | English |
| last_indexed | 2024-03-09T03:17:21Z |
| publishDate | 2021-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Journal of Imaging |
| spelling | doaj.art-d5188fd600da4d0ba25183fad6515e2e2023-12-03T15:18:30ZengMDPI AGJournal of Imaging2313-433X2021-01-01722010.3390/jimaging7020020Improved Visual Localization via Graph FilteringCarlos Lassance0Yasir Latif1Ravi Garg2Vincent Gripon3Ian Reid4Electronics Department, IMT Atlantique, 29280 Brest, FranceSchool of Computer Science, University of Adelaide, Adelaide 5005, AustraliaSchool of Computer Science, University of Adelaide, Adelaide 5005, AustraliaElectronics Department, IMT Atlantique, 29280 Brest, FranceSchool of Computer Science, University of Adelaide, Adelaide 5005, AustraliaVision-based localization is the problem of inferring the pose of the camera given a single image. One commonly used approach relies on image retrieval where the query input is compared against a database of localized support examples and its pose is inferred with the help of the retrieved items. This assumes that images taken from the same places consist of the same landmarks and thus would have similar feature representations. These representations can learn to be robust to different variations in capture conditions like time of the day or weather. In this work, we introduce a framework which aims at enhancing the performance of such retrieval-based localization methods. It consists in taking into account additional information available, such as GPS coordinates or temporal proximity in the acquisition of the images. More precisely, our method consists in constructing a graph based on this additional information that is later used to improve reliability of the retrieval process by filtering the feature representations of support and/or query images. We show that the proposed method is able to significantly improve the localization accuracy on two large scale datasets, as well as the mean average precision in classical image retrieval scenarios.https://www.mdpi.com/2313-433X/7/2/20graph signal processingdeep learningvisual localizationimage retrievaltransfer learning |
| spellingShingle | Carlos Lassance Yasir Latif Ravi Garg Vincent Gripon Ian Reid Improved Visual Localization via Graph Filtering Journal of Imaging graph signal processing deep learning visual localization image retrieval transfer learning |
| title | Improved Visual Localization via Graph Filtering |
| title_full | Improved Visual Localization via Graph Filtering |
| title_fullStr | Improved Visual Localization via Graph Filtering |
| title_full_unstemmed | Improved Visual Localization via Graph Filtering |
| title_short | Improved Visual Localization via Graph Filtering |
| title_sort | improved visual localization via graph filtering |
| topic | graph signal processing deep learning visual localization image retrieval transfer learning |
| url | https://www.mdpi.com/2313-433X/7/2/20 |
| work_keys_str_mv | AT carloslassance improvedvisuallocalizationviagraphfiltering AT yasirlatif improvedvisuallocalizationviagraphfiltering AT ravigarg improvedvisuallocalizationviagraphfiltering AT vincentgripon improvedvisuallocalizationviagraphfiltering AT ianreid improvedvisuallocalizationviagraphfiltering |