TU4.R12: Deep Learning Methods for Multispectral Image Analysis II

Session Type: Oral
Time: Tuesday, July 24, 16:50 - 18:30
Location: Room 2F
Session Chairs: Matthieu Molinier, VTT and Mihai Datcu, DLR
 
16:50 - 17:10
TU4.R12.1: IMPROVING MAPS FROM CNNS TRAINED WITH SPARSE, SCRIBBLED GROUND TRUTHS USING FULLY CONNECTED CRFS
         Luca Maggiolo; University of Genoa
         Diego Marcos; Wageningen University
         Gabriele Moser; University of Genoa
         Devis Tuia; Wageningen University
 
17:10 - 17:30
TU4.R12.2: CONVOLUTIONAL NEURAL NETWORKS FOR CLOUD SCREENING: TRANSFER LEARNING FROM LANDSAT-8 TO PROBA-V
         Gonzalo Mateo-García; Universidad de València
         Luis Gómez-Chova; Universidad de València
 
17:30 - 17:50
TU4.R12.3: DEEPCLOUD - A FULLY CONVOLUTIONNAL NEURAL NETWORK FOR CLOUD AND SHADOW MASKING IN OPTICAL SATELLITE IMAGES
         Matthieu Molinier; VTT Technical Research Centre of Finland Ltd
         Niko Reunanen; VTT Technical Research Centre of Finland Ltd
         Arttu Lämsä; VTT Technical Research Centre of Finland Ltd
         Heikki Astola; VTT Technical Research Centre of Finland Ltd
         Tomi Räty; VTT Technical Research Centre of Finland Ltd
 
17:50 - 18:10
TU4.R12.4: TRANSFER LEARNING WITH CONVOLUTIONAL NETWORKS FOR ATMOSPHERIC PARAMETER RETRIEVAL
         David Malmgren-Hansen; Danmarks Tekniske Universitet
         Valero Laparra; University of Valencia
         Allan Aasbjerg Nielsen; Danmarks Tekniske Universitet
         Gustau Camps-Valls; University of Valencia
 
18:10 - 18:30
TU4.R12.5: URBAN CHANGE DETECTION FOR MULTISPECTRAL EARTH OBSERVATION USING CONVOLUTIONAL NEURAL NETWORKS
         Rodrigo Caye Daudt; ONERA
         Bertrand Le Saux; ONERA
         Alexandre Boulch; ONERA
         Yann Gousseau; Télécom ParisTech