Abstract

Accurate agricultural field boundary delineation at large scale is a foundational task for food security, supply chain transparency, and carbon accounting. While vision foundation models like SAM show remarkable zero-shot capabilities, they frequently fail in geospatial domains due to topological complexity, cropland texturing patterns, and a lack of physical scale awareness. In this work, we introduce Delineate Anything v2, a globally scalable foundation model designed specifically for wide-area field boundary mapping. We construct FBIS-73M, a 73-million-instance multi-resolution dataset spanning 61 countries. To address the pervasive issue of multi-field administrative parcel merging, we introduce a resolution-specific data curation pipeline that leverages topological image-space adaptation to homogenize merged parcels and strengthen weak physical boundaries. Furthermore, we establish a novel, manually curated evaluation benchmark covering 100 countries to assess independent zero-shot generalization. Our results show that Delineate Anything v2 surpasses the current state-of-the-art, including the Delineate Anything framework, by 0.284 mAP@0.5 (+103.3% relative gain), while maintaining execution speeds suitable for rapid national- and global-scale deployment, as demonstrated by nationwide mapping of Ukraine (603,000 km²) in 5.4 hours on a consumer-grade workstation.

Methodology

1. A novel task formulation of field boundary detection as an instance segmentation problem, addressing the inherent limitations of semantic segmentation for this task.
2. We construct FBIS-73M, a 73-million-instance multi-resolution field boundary repository spanning 61 countries — the largest public dataset of its kind.
3. We propose a data-centric noise remediation strategy that resolves the parcel-merging anomaly by combining manual partitioning for HR data with automated image-space pixel homogenization for MR data.
4. We manually curate a high-fidelity validation benchmark spanning 100 countries to evaluate generalization across diverse agricultural regimes.
5. We demonstrate that Delineate Anything v2 establishes a new state-of-the-art for global agricultural field delineation, surpassing the previous state-of-the-art Delineate Anything framework by 0.284 mAP@0.5 (+103.3% relative gain) and enabling nationwide mapping of Ukraine (603,000 km²) in 5.4 hours on a consumer-grade workstation.

Field Boundary Instance Segmentation - 73M (FBIS-73M) dataset

FBIS-73M extends FBIS-22M into a 73-million-instance, multi-resolution (0.25m–10m) dataset spanning 61 countries with diverse imagery sources, built through a resolution-specific curation pipeline that resolves the parcel-versus-field mismatch present in raw administrative boundary sources.

(a) Training set (FBIS-73M)

(b) Evaluation benchmark (100 countries)

Results

Note: For a quantitative and qualitative comparison of Delineate Anything against generic vision foundation models (SAM, SAM2) and MultiTLF, see Table 2 and Figure 4 in the Delineate Anything paper.

Qualitative Visualization

BibTeX

@inproceedings{lavreniuk2026delanyv2,
  title={Delineate Anything v2: A Global Foundation Model for Field Delineation},
  author={Mykola Lavreniuk and Nataliia Kussul and Andrii Shelestov and Yevhenii Salii and Volodymyr Kuzin and Charlotte Julia Li-Xing Wang and Zoltan Szantoi},
  year={2026},
  booktitle={European Conference on Computer Vision Workshops (ECCVW)},
}

@inproceedings{lavreniuk2025delineateanything,
  title={Delineate Anything: Resolution-Agnostic Field Boundary Delineation on Satellite Imagery},
  author={Mykola Lavreniuk and Nataliia Kussul and Andrii Shelestov and Bohdan Yailymov and Yevhenii Salii and Volodymyr Kuzin and Zoltan Szantoi},
  year={2025},
  booktitle={European Conference on Artificial Intelligence},
}

@article{lavreniuk2025delineateanythingflow,
  title={Delineate Anything Flow: Fast, Country-Level Field Boundary Detection from Any Source},
  author={Mykola Lavreniuk and Nataliia Kussul and Andrii Shelestov and Yevhenii Salii and Volodymyr Kuzin and Sergii Skakun and Zoltan Szantoi},
  year={2025},
  journal={https://arxiv.org/abs/2511.13417},
}