We are excited to announce our 2nd workshop on Complex Data Challenges in Earth Observation (CDCEO). Submission of contributions is now open. CDCEO is part of the 31st International Joint Conference on Artificial Intelligence and the 25th European Conference on Artificial Intelligence (IJCAI-ECAI) running on July 23-29, 2022 in Vienna, Austria. The workshop is organized in collaboration between IARAI, The European Space Agency, The University of Tokyo, Griffith University, Microsoft AI for Good, Maxar Technologies, and The Geoscience and Remote Sensing Society.
Modern remote sensing technologies allow monitoring our planet at an unprecedented scale and resolution. A multitude of ground, aerial, and satellite-based sensors collect a wealth of observation data. Effective extraction and interpretation of such information requires powerful data-driven algorithms. Remote sensing data vary in spatiotemporal scales, collection frequencies, spectral ranges, and dimensionality and have diverse application domains. Complexity, volume, and multi-scale and multi-source character of Earth observation data present unique challenges for algorithm development.
CDCEO 2022 aims to advance research in Earth observation using cutting-edge AI algorithms. The workshop topics are not limited to specific domains or data sources. This cross-disciplinary workshop brings together researchers in AI, big data, remote sensing, geographic information systems, weather and climate modeling, and other domains. We invite contributions in both applications and method development in image and signal processing, feature extraction, data fusion, meta-learning, and many more.
CDCEO 2022 will feature benchmark datasets for training and testing AI models. We will introduce the first dataset with transferrable adversarial examples in remote sensing. A special session of the workshop will showcase the results of the Landslide4Sense competition. Landslide4Sense focuses on innovative algorithms for landslide detection using remote sensing data. The competition provides a benchmark dataset of globally distributed multi-sensor satellite images.
