Fields of The World: A Machine Learning Benchmark Dataset For Global Agricultural Field Boundary Segmentation

Paper

We provide results from baseline models for the new Fields of The World (FTW) benchmark, show that models trained on FTW have better zero-shot and fine-tuning performance in held-out countries than models that are not pre-trained with diverse datasets, and show positive qualitative zero-shot results of FTW models in a real-world scenario – running on Sentinel-2 scenes over Ethiopia.

Figure 1. Training samples from four continents, demonstrating the diversity within Fields of The World.

Cite as:

@article{kerner2024fields,
  title={Fields of The World: A Machine Learning Benchmark Dataset For Global Agricultural Field Boundary Segmentation},
  author={Kerner, Hannah and Chaudhari, Snehal and Ghosh, Aninda and Robinson, Caleb and Ahmad, Adeel and Choi, Eddie and Jacobs, Nathan and Holmes, Chris and Mohr, Matthias and Dodhia, Rahul and others},
  journal={arXiv preprint arXiv:2409.16252},
  year={2024}
}