5 papers
Reuse or Relearn? A Spectral View of Earth Observation Foundation Models
Mehmet Ozgur Turkoglu, Valerio Marsocci, Dominik J. Mühlematter +3
Foundation models are rarely used as generic, frozen feature extractors; instead, they are fine-tuned for the target downstream application. This practice is particularly prevalent…
Towards Robust Deep Learning-based Rumex Obtusifolius Detection from Drone Images
Fabian Dionys Schrag, Mehmet Ozgur Turkoglu, Konrad Schindler +1
Domain adaptation (DA) addresses the challenge of transferring a machine learning model trained on a source domain to a target domain with a different data distribution. In this wo…
TerraFlow: Multimodal, Multitemporal Representation Learning for Earth Observation
Nazar Puriy, Johannes Jakubik, Benedikt Blumenstiel +1
We propose TerraFlow, a novel approach to multimodal, multitemporal learning for Earth observation. TerraFlow builds on temporal training objectives that enable sequence-aware lear…
TerraCodec: Compressing Optical Earth Observation Data
Julen Costa-Watanabe, Isabelle Wittmann, Benedikt Blumenstiel +1
Earth observation (EO) satellites produce massive streams of multispectral image time series, posing pressing challenges for storage and transmission. Yet, learned EO compression r…
Fine-tune Smarter, Not Harder: Parameter-Efficient Fine-Tuning for Geospatial Foundation Models
Francesc Marti-Escofet, Benedikt Blumenstiel, Linus Scheibenreif +2
Earth observation (EO) is crucial for monitoring environmental changes, responding to disasters, and managing natural resources. In this context, foundation models facilitate remot…