Eiffel Tower: A Deep-Sea Underwater Dataset for Long-Term Visual Localization

Type Article
Date 2023-08
Language English
Author(s) Boittiaux Clementin1, 2, 3, Dune-Maillard Claire2, Ferrera MaximeORCID1, Arnaubec Aurelien1, Marxer Ricard3, Matabos MarjolaineORCID4, Van Audenhaege LoicORCID4, Hugel Vincent2
Affiliation(s) 1 : Ifremer, Zone Portuaire de Bregaillon, La Seyne-sur-Mer, France
2 : Universite de Toulon, COSMER, Toulon, France
3 : Universite de Toulon, Aix Marseille Univ, CNRS, LIS, Toulon, France
4 : Univ Brest, CNRS, Ifremer, UMR6197 BEEP, F-29280 Plouzane, France
Source International Journal Of Robotics Research (0278-3649) (SAGE Publications), 2023-08 , Vol. 42 , N. 9 , P. 689-699
DOI 10.1177/02783649231177322
WOS© Times Cited 5
Note Data paper
Keyword(s) Underwater dataset, long-term visual localization, deep sea, visual localization benchmark, Eiffel Tower vent edifice
Abstract

Visual localization plays an important role in the positioning and navigation of robotics systems within previously visited environments. When visits occur over long periods of time, changes in the environment related to seasons or daynight cycles present a major challenge. Under water, the sources of variability are due to other factors such as water conditions or growth of marine organisms. Yet it remains a major obstacle and a much less studied one, partly due to the lack of data. This paper presents a new deep-sea dataset to benchmark underwater long-term visual localization. The dataset is composed of images from four visits to the same hydrothermal vent edifice over the course of five years. Camera poses and a common geometry of the scene were estimated using navigation data and Structure-from-Motion. This serves as a reference when evaluating visual localization techniques. An analysis of the data provides insights about the major changes observed throughout the years. Furthermore, several well-established visual localization methods are evaluated on the dataset, showing there is still room for improvement in underwater long-term visual localization. The data is made publicly available at seanoe.org/data/00810/92226/.

Full Text
File Pages Size Access
Preprint - 10.48550/arXiv.2305.05301 10 10 MB Open access
11 2 MB Access on demand
Top of the page

How to cite 

Boittiaux Clementin, Dune-Maillard Claire, Ferrera Maxime, Arnaubec Aurelien, Marxer Ricard, Matabos Marjolaine, Van Audenhaege Loic, Hugel Vincent (2023). Eiffel Tower: A Deep-Sea Underwater Dataset for Long-Term Visual Localization. International Journal Of Robotics Research, 42(9), 689-699. Publisher's official version : https://doi.org/10.1177/02783649231177322 , Open Access version : https://archimer.ifremer.fr/doc/00836/94826/