Stacked Encoder-Decoders for Accurate Semantic Segmentation of Very High Resolution Satellite Datasets

Abstract : Semantic segmentation is a mainstream method in several remote sensing applications based on very-high-resolution data, achieving recently remarkable performance by the use of deep learning and more specifically, pixel-wise dense classification models. In this paper, we exploit the use of a relatively deep architecture based on repetitive downscale-upscale processes that had been previously employed for human pose estimation. By integrating such a model, we are aiming to capture low-level details, such as small objects, object boundaries and edges. Experimental results and quantitative evaluation has been performed on the publicly available ISPRS (WGIII/4) benchmark dataset indicating the potential of the proposed approach.
Type de document :
Communication dans un congrès
IGARSS 2018 - 38th annual International Geoscience and Remote Sensing Symposium, Jul 2018, Valencia, Spain. pp.1-4
Liste complète des métadonnées

https://hal.archives-ouvertes.fr/hal-01870857
Contributeur : Maria Papadomanolaki <>
Soumis le : dimanche 9 septembre 2018 - 21:49:06
Dernière modification le : mardi 2 octobre 2018 - 01:15:38

Fichier

IGARSS2018_v8.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-01870857, version 1

Citation

Maria Papadomanolaki, Maria Vakalopoulou, Nikos Paragios, Konstantinos Karantzalos. Stacked Encoder-Decoders for Accurate Semantic Segmentation of Very High Resolution Satellite Datasets. IGARSS 2018 - 38th annual International Geoscience and Remote Sensing Symposium, Jul 2018, Valencia, Spain. pp.1-4. 〈hal-01870857〉

Partager

Métriques

Consultations de la notice

76

Téléchargements de fichiers

87