Machine Learning for Earth Observation

In Conjunction with the ECML/PKDD 2021

Virtual, September 17, 2021

The workshop will be fully virtual via Zoom platform.

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

FULL PROGRAM - Friday, September 17, 2021

9h15 - 9h30 Welcome

9h30 - 10h45 Invited Talk: Prof. Begüm Demir, Deep Earth Query: Information Discovery from Big Earth Observation Data Archives [slides]

10h45 - 11h00 Break

11h00 - 12h00 First Session

  • Christel Chappuis, Sylvain Lobry, Benjamin Kellenberger, Bertrand Le Saux and Devis Tuia. How to find a good image-text embedding for remote sensing visual question answering? [paper] [slides]

  • Sudipan Saha, Lichao Mou, Muhammad Shahzad and Xiao Xiang Zhu. Segmentation of VHR EO Images using Unsupervised Learning [paper]


14h30 - 15h30 Second Session

  • Mihailo Obrenović, Thomas Lampert, François Monde-Kossi, Miloš Ivanovic and Pierre Gançarski. SS-HIDA: Semi-Supervised Heterogeneous Image Domain Adaptation [paper] [slides]

  • Matthew Gibson and Arcot Sowmya. Semi-supervised Siamese Networks for Change Recognition with VHR Imagery [paper] [slides]

15h30 - 15h45 Break

15h45 - 17h00 Prof. Gustau Camps-Valls, Fitting is not enough: Interpretability and causal inference for the Earth sciences [slides]

17h00 - 17h15 Discussion and Closing