Hedgementation: A Remote Sensing Benchmark for Hedgerow Segmentation

The authors propose Hedgementation, a new benchmark designed to evaluate machine learning models for mapping hedgerows from remote sensing data at a country scale with 10m² spatial resolution. This initiative combines and harmonizes multiple remote sensing products and ground truth labels derived from a French hedgerow inventory.

• The benchmark assesses model generalization across spatial distances and climatic zones.
• It tests both supervised and self-supervised learning approaches for tracking fine-scale agricultural features.
• Reproducible code and baseline results are available on GitHub.

This resource supports the development of models capable of accurately monitoring high-importance agricultural infrastructure using diverse remote sensing inputs.