Sep 2026 – now · Gainesville, FL
GeoDI Lab, University of Florida
Research assistant, geospatial machine learning
I work on a project about filling gaps in satellite imagery: predicting a sharp image for days when only a coarse one was taken, and testing how well large pretrained models for satellite data (geospatial foundation models) cope when some of a sensor's colour or infrared channels are missing.
My part is reproducing the baseline model, running experiments, comparing it with existing methods and analysing the results.
Python · PyTorch · Remote sensing
Read the project write-upDiagram: a fine-resolution image from an earlier date and a coarse image from the target date are fused to predict a fine image at the target date, with one spectral band withheld.