Semantic Segmentation of Earth Observation Data Using Multimodal and Multi-scale Deep Networks

Nicolas Audebert 1, 2 Bertrand Le Saux 1 Sébastien Lefèvre 2
2 OBELIX - Environment observation with complex imagery
UBS - Université de Bretagne Sud, IRISA-D5 - SIGNAUX ET IMAGES NUMÉRIQUES, ROBOTIQUE
Abstract : This work investigates the use of deep fully convolutional neural networks (DFCNN) for pixel-wise scene labeling of Earth Observation images. Especially, we train a variant of the SegNet architecture on remote sensing data over an urban area and study different strategies for performing accurate semantic segmentation. Our contributions are the following: 1) we transfer efficiently a DFCNN from generic everyday images to remote sensing images; 2) we introduce a multi-kernel convolutional layer for fast aggregation of predictions at multiple scales; 3) we perform data fusion from heterogeneous sensors (optical and laser) using residual correction. Our framework improves state-of-the-art accuracy on the ISPRS Vaihingen 2D Semantic Labeling dataset.
Type de document :
Communication dans un congrès
Asian Conference on Computer Vision (ACCV16), Nov 2016, Taipei, Taiwan. <http://www.accv2016.org/>
Liste complète des métadonnées


https://hal.archives-ouvertes.fr/hal-01360166
Contributeur : Nicolas Audebert <>
Soumis le : mardi 20 septembre 2016 - 16:05:39
Dernière modification le : samedi 18 février 2017 - 01:09:40

Fichiers

accv16_final_483.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-01360166, version 1
  • ARXIV : 1609.06846

Citation

Nicolas Audebert, Bertrand Le Saux, Sébastien Lefèvre. Semantic Segmentation of Earth Observation Data Using Multimodal and Multi-scale Deep Networks. Asian Conference on Computer Vision (ACCV16), Nov 2016, Taipei, Taiwan. <http://www.accv2016.org/>. <hal-01360166>

Partager

Métriques

Consultations de
la notice

814

Téléchargements du document

230