Workshop on Machine Learning for Earth Observation
In Conjunction with the ECML/PKDD 2024
Vilnius, September 13, 2024
FULL PROGRAM - Friday, September 13/09/2024
09:00 - 09:15 Welcome
09:15 - 10:00 Invited Talk 1: Matthieu Molinier, VTT Technical Research Centre of Finland, Improving temporal frequency of dense multispectral image time series applications for environmental monitoring
10:00 - 11:00 Session 1: Feature selection and dimensionality reduction
Leveraging band diversity for feature selection in EO data
Sadia Hussain (IIT Delhi)*; Brejesh Lall (IIT Delhi)Data-Centric Machine Learning for Earth Observation: Necessary and Sufficient Features
Hiba Najjar (University of Kaiserslautern-Landau & DFKI GmbH)*; Marlon Nuske (DFKI GmbH); Andreas Dengel (DFKI GmbH)Dimensionality Reduction for Efficient Semi-supervised Learning from Remote Sensing Images
Sintija Stevanoska (Jožef Stefan International Postgraduate School & Jožef Stefan Institute)*; Marjan Stoimchev (Jožef Stefan Institute), Jurica Levatić (Jožef Stefan Institute), Sašo Džeroski (Jožef Stefan Institute)
11:00 - 11:30 Coffee Break
11:30 - 12:30 Session 2: Multi-modal and multi-label settings
Increasing the Robustness of Model Predictions to Missing Sensors in Earth Observation
Francisco Mena (University of Kaiserslautern-Landau & DFKI GmbH)*; Diego Arenas (German Research Center for Artificial Intelligence); Andreas Dengel (University of Kaiserslautern-Landau & DFKI GmbH)Domain generalized remote sensing scene captioning via Wikipedia geography articles
Kerem Aydın (Sabanci University)*; Erchan Aptoula (Sabanci University)Hierarchical Multi-label Classification of Land Use/ Land Cover in Remote Sensing Images
Marjan Stoimchev*(Jožef Stefan International Postgraduate School & Jožef Stefan Institute), Jurica Levatić, Dragi Kocev, and Sašo Džeroski (Jožef Stefan Institute)
12:30 - 14:00 Lunch break
14:00 - 14:45 Invited Talk 2: Giuseppina Andresini & Annalisa Appice, Università di Bari, Italy, Monitoring forest health with AI: approaches for mapping tree dieback in satellite data
14:45 - 15:30 Session 3: Detection and Monitoring
Estimating Earthquake Magnitude in Sentinel-1 Imagery via Ranking
Daniele Rege Cambrin (Politecnico di Torino); Isaac A Corley (University of Texas at San Antonio)*; Paolo Garza (Politecnico di Torino); Peyman Najafirad (University of Texas at San Antonio)A Clustering-based Algorithm for Fast Anomaly Detection in Hyperspectral Images
Mark Last (Ben-Gurion University of the Negev)*A Deep Learning-Based Approach for Mangrove Monitoring
Lucas José Velôso de Souza (CentraleSupelec); Ingrid Valverde Reis Zreik (CentraleSupélec); Adrien Salem-Sermanet (CentraleSupélec); Nacéra Seghouani (LISN - Paris-Saclay University)*; Lionel Pourchier (EVERSEA)
15:30 Closing