Nico Lang
Nico is an Assistant Professor at the University of Copenhagen associated with the Global Wetland Centre and the Pioneer Centre for AI. He is also a core member of the Climate AI Nordics network. Before moving to Denmark for a Postdoc, Nico has received a PhD from ETH Zurich. His research focuses on computer vision, machine learning, and remote sensing, and on developing new methods to support environmental sciences. More information can be found on his website: https://langnico.github.io/
Sessions
The volume of unlabeled Earth observation (EO) data is huge. To interpret this vast amount of data, efficient modelling approaches are needed that can generalize to large geographic areas and are robust to inherent noise. Data-driven approaches promise great potential for interpreting and combining data from different space missions. In this talk, I will present our work on global canopy height mapping (https://langnico.github.io/globalcanopyheight/) with optical satellite images and sparse spaceborne lidar data and discuss a recent project called MMEarth (https://vishalned.github.io/mmearth/) that explored multi-modal pretext tasks for learning representations that are suitable for a range of downstream tasks with limited training data.
Machine learning has become an important toolbox for analyzing complex Earth observation data to derive information from the raw data. In particular, supervised deep learning has achieved great success in solving EO tasks where the relationship between input and output is not clearly understood. However, applications with limited reference data cannot directly benefit from advances in supervised deep learning. This lecture will first introduce the concepts of supervised deep learning and then provide an overview of research in the field of self-supervised learning (SSL), which aims to learn transferable representations (i.e., features) from unlabeled data.