New article in pre-print titled "Taking it further; leveraging pseudo labels for field delineation across label-scarce smallholder regions"

Dr. Mirela Tulbure contributed to a new article that is currently in pre-print entitled “Taking it further; leveraging pseudo labels for field delineation across label-scarce smallholder regions.”

The lead author of this article is Dr. Philippe Rufin of the Earth and Life Institute at UCLouvain and the Geography department at Humboldt University. From the abstract, “This study explores opportunities for using pre-trained models to generate sparse (i.e. not fully annotated) field delineation pseudo labels for fine-tuning models across geographies and sensor characteristics. We build on a FracTAL ResUNet trained for crop field delineation in India (median field size of 0.24 ha) based on multi-spectral imagery at 1.5 m spatial resolution”

The full article is available here.

Social Media Buzz:

Smallholder field delineation based on #earthobservation & #deeplearning is challenged by a need for more training data.

Pseudo-labels support domain adaptation across geographies & sensors.

Pre-print led by Dr. Rufin https://t.co/BT4rUWySHr #computervision #eochat pic.twitter.com/XpIUx5TieY

— Dr. Mirela G. Tulbure 🛰 🌎 + 🐍 + 🌊 (@MirelaGTulbure) December 18, 2023

Bibliographic Citation

Rufin, P., Wang, S., Lisboa, S. N., Hemmerling, J., Tulbure, M. G., Meyfroidt, P. (2023). Taking it further: leveraging pseudo labels for field delineation across label-scarce smallholder regions. https://doi.org/10.48550/arXiv.2312.08384