Machine Learning for Earth Observation

In Conjunction with the ECML/PKDD 2020

Ghent, Belgium, September 18 2020

=========================================

PAPER AWARDS

Congratulations to the winners of the MACLEAN paper awards sponsored by ESA (European Space Agency).

  • Best Paper Award

Vit Ruzicka, Stefano D'Aronco, Jan Dirk Wegner and Konrad Schindler. Deep Active Learning in Remote Sensing for data efficient Change Detection

  • Best Student Paper Award

Gaston Lenczner. Interactive Learning for Semantic Segmentation in Earth Observation

=========================================

The workshop will be fully virtual via Zoom platform.

Access link to the meeting room: https://zoom.us/j/92102396066?pwd=WnF2Rnp6by85VlVjNk8yZHI3KzBaQT09


FULL PROGRAM - September 18, 2020 (NOTE: this is in CEST=UTC+2)

9h00 - 9h15 Welcome

9h15 - 10h15 Invited Speaker: Prof. Dr. Ir. Aleksandra Pizurica, Sparse coding and deep learning in the analysis of hyperspectral images in remote sensing (link to keynote)


10h30 - 11h30 First Session: Semi-Supervised and (Inter)active Learning

  • Gaston Lenczner, Adrien Chan-Hon-Tong, Nicola Luminari, Bertrand Le Saux and Guy Le Besnerais. Interactive Learning for Semantic Segmentation in Earth Observation (paper)

  • Vit Ruzicka, Stefano D'Aronco, Jan Dirk Wegner and Konrad Schindler. Deep Active Learning in Remote Sensing for data efficient Change Detection (paper)

  • Javiera Castillo-Navarro, Bertrand Le Saux, Alexandre Boulch and Sébastien Lefèvre. On Auxiliary Losses for Semi-Supervised Semantic Segmentation (paper)


11h45 - 12h45 Second Session: Physical event characterization

  • Yousef Razeghi, Wilhelm Hasselbring, Christian Berndt and Ines Dumke. Comparing Machine Learning Methods for Predicting Seismic P-wave Velocity on Global Scale (paper)

  • Clémentin Boittiaux, Paul Nguyen Hong Duc and Dorian Cazau. Embedding underwater passive acoustics and SAR images in multi-modal deep generative models for wind speed classification (paper)

  • Abhijit Mukherjee and Pabitra Mitra. Joint Learning for Cyclone Track Nowcasting (paper)


14h15 - 15h15 Invited Speaker: Prof. Devis Tuia, Making sense of geospatial data with deep learning … while interacting with people (link to keynote)


15h30 - 16h30 Third Session: Multi-spectral and Colorization

  • Youssef Mourchid, Marc Donias and Yannick Berthoumieu. Dual Color-Image Discriminators Adversarial Networks for Generating Artificial-SAR Colorized Images from SENTINEL-1 Images (paper)

  • Pallavi Jain, Bianca Schoen-Phelan and Robert Ross. Tri-Band Assessment of Multi-Spectral Satellite Data for Flood Detection (paper)

  • Estèle Glize, Marie-José Huguet, Gilles Tredan, Marc Lucas and Marion Sutton. Clustering Sargassum Mats from Earth Observation Data (paper)


16h30 - 17h Discussion and Closing