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

Date 2022
Temporal extent 2015 -2020
Author(s) Boittiaux ClementinORCID1, 2, 3, Dune ClaireORCID2, Ferrera MaximeORCID1, Arnaubec AurelienORCID1, Marxer RicardORCID3, Van Audenhaege LoicORCID4, Matabos MarjolaineORCID4, Hugel VincentORCID2
Affiliation(s) 1 : Ifremer, Zone Portuaire de Brégaillon, La Seyne-sur-Mer, France
2 : Université de Toulon, COSMER, Toulon, France
3 : Université de Toulon, Aix Marseille Univ, CNRS, LIS, Toulon, France
4 : Univ Brest, CNRS, Ifremer, UMR6197 BEEP, F-29280 Plouzané, France
DOI 10.17882/92226
Publisher SEANOE
Keyword(s) Eiffel Tower, underwater dataset, visual localization, deep sea
Abstract

Visual localization plays an important role in 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 day/night 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.

Licence CC-BY-NC-ND
Acknowledgements All the acquisitions were conducted by Ifremer. The authors of the dataset would like to thank the crews of the research vessels Pourquoi Pas? and L'Atalante, the pilots of the ROV Victor6000 as well as all the personnel who helped in acquiring this data.
Sensor metadata

Full HD images extracted from ROV Victor6000 HD and 4k cameras.

Data
File Size Format Processing Access
2015 images, navigation and model 14 GB ZIP Processed data Open access
2016 images, navigation and model 10 GB ZIP Processed data Open access
2018 images, navigation and model 15 GB ZIP Processed data Open access
2020 images, navigation and model 9 GB ZIP Processed data Open access
Global model 287 MB ZIP Processed data Open access
Top of the page

How to cite 

Boittiaux Clementin, Dune Claire, Ferrera Maxime, Arnaubec Aurelien, Marxer Ricard, Van Audenhaege Loic, Matabos Marjolaine, Hugel Vincent (2022). Eiffel Tower: A Deep-Sea Underwater Dataset for Long-Term Visual Localization. SEANOE. https://doi.org/10.17882/92226