Fusion of Heterogeneous Data in Convolutional Networks for Urban Semantic Labeling (Invited Paper) - Archive ouverte HAL Access content directly
Conference Papers Year : 2017

Fusion of Heterogeneous Data in Convolutional Networks for Urban Semantic Labeling (Invited Paper)

Abstract

In this work, we present a novel module to perform fusion of heterogeneous data using fully convolutional networks for semantic labeling. We introduce residual correction as a way to learn how to fuse predictions coming out of a dual stream architecture. Especially, we perform fusion of DSM and IRRG optical data on the ISPRS Vaihingen dataset over a urban area and obtain new state-of-the-art results.
Fichier principal
Vignette du fichier
audebert_lesaux_lefevre_data_fusion_for_urban_remote_sensing.pdf (661.45 Ko) Télécharger le fichier
correct_network.pdf (17.69 Ko) Télécharger le fichier
dual_segnet_fusion.pdf (13.4 Ko) Télécharger le fichier
dual_segnet_sum.pdf (10.47 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Origin : Files produced by the author(s)
Origin : Files produced by the author(s)
Origin : Files produced by the author(s)

Dates and versions

hal-01438499 , version 1 (17-01-2017)

Identifiers

Cite

Nicolas Audebert, Bertrand Le Saux, Sébastien Lefèvre. Fusion of Heterogeneous Data in Convolutional Networks for Urban Semantic Labeling (Invited Paper). Joint Urban Remote Sensing Event (JURSE), Mar 2017, Dubai, United Arab Emirates. ⟨10.1109/JURSE.2017.7924566⟩. ⟨hal-01438499⟩
581 View
546 Download

Altmetric

Share

Gmail Facebook X LinkedIn More