Synthetic Datasets and Evaluations for Sub-pixel Displacements Estimation from Optical Satellite Images with Deep Learning

Dataset

DOI

Contains 3 synthetic datasets (UNI, DIS, UNI-5px), with three corresponding trained models, based on a CNN architecture. The three datasets contains pairs of small (either 16x16 or 32x32 pixels) windows that simulate shifts.

Contains also evaluations on realistic synthetics examples of a deep learning pipeline using two of the three models presented.

Identifier
DOI	https://doi.org/10.57745/UOGRPY
Metadata Access	https://entrepot.recherche.data.gouv.fr/oai?verb=GetRecord&metadataPrefix=oai_datacite&identifier=doi:10.57745/UOGRPY

Provenance
Creator	Montagnon, Tristan
Publisher	Recherche Data Gouv
Contributor	Montagnon, Tristan; James Hollingsworth; Hollingsworth, James; Giffard-Roisin, Sophie; Dalla Mura, Mauro; Marchandon, Mathilde; Pathier, Erwan; Institut des Sciences de la Terre; Entrepôt-Catalogue Recherche Data Gouv
Publication Year	2023
Rights	etalab 2.0; info:eu-repo/semantics/openAccess; https://spdx.org/licenses/etalab-2.0.html
OpenAccess	true
Contact	Montagnon, Tristan (ISTerre ; UGA, IRD, CNRS, USMB, Université Gustave Eiffel ; France); James Hollingsworth (ISTerre ; UGA, IRD, CNRS, USMB, Université Gustave Eiffel ; France)

Representation
Resource Type	Dataset
Format	application/zip; text/plain
Size	1517225814; 83023495; 17882571; 3638
Version	1.0
Discipline	Geosciences; Earth and Environmental Science; Environmental Research; Geology; Geospheric Sciences; Natural Sciences