Skip to Main content Skip to Navigation
Conference papers

What Data are needed for Semantic Segmentation in Earth Observation?

Abstract : This paper explores different aspects of semantic segmentation of remote sensing data using deep neural networks. Learning with deep neural networks was revolutionized by the creation of ImageNet. Remote sensing benefited of these new techniques, however Earth Observation (EO) datasets remain small in comparison. In this work, we investigate how we can progress towards the ImageNet of remote sensing. In particular, two questions are addressed in this paper. First, how robust are existing supervised learning strategies with respect to data volume? Second, which properties are expected from a large-scale EO dataset? The main contributions of this work are: (i) a strong robustness analysis of existing supervised learning strategies with respect to remote sensing data, (ii) the introduction of a new, large-scale dataset named MiniFrance.
Document type :
Conference papers
Complete list of metadatas

Cited literature [17 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-02343915
Contributor : Sébastien Lefèvre <>
Submitted on : Tuesday, December 10, 2019 - 8:04:14 AM
Last modification on : Thursday, January 7, 2021 - 4:39:22 PM
Long-term archiving on: : Wednesday, March 11, 2020 - 1:29:18 PM

File

castillo_what-data_jurse19_pre...
Files produced by the author(s)

Identifiers

Citation

Javiera Castillo-Navarro, Nicolas Audebert, Alexandre Boulch, Bertrand Le Saux, Sébastien Lefèvre. What Data are needed for Semantic Segmentation in Earth Observation?. 2019 Joint Urban Remote Sensing Event (JURSE), May 2019, Vannes, France. pp.1-4, ⟨10.1109/JURSE.2019.8809071⟩. ⟨hal-02343915v2⟩

Share

Metrics

Record views

190

Files downloads

446