Semantic Segmentation of Earth Observation Data Using Multimodal and Multi-scale Deep Networks - Archive ouverte HAL Access content directly
Conference Papers Year : 2016

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

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.
Fichier principal
Vignette du fichier
accv16_final_483.pdf (1.14 Mo) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-01360166 , version 1 (20-09-2016)

Identifiers

Cite

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. ⟨10.1007/978-3-319-54181-5_12⟩. ⟨hal-01360166⟩
1157 View
566 Download

Altmetric

Share

Gmail Facebook X LinkedIn More