Pith. sign in

REVIEW 9 cited by

Continental-Scale Building Detection from High Resolution Satellite Imagery

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2107.12283 v2 pith:7YFGFZ74 submitted 2021-07-26 cs.CV

Wojciech Sirko , Sergii Kashubin , Marvin Ritter , Abigail Annkah , Yasser Salah Eddine Bouchareb , Yann Dauphin , Daniel Keysers , Maxim Neumann
show 2 more authors
Moustapha Cisse John Quinn
This is my paper · ORCID
classification cs.CV
keywords satellitebuildingbuildingsmodelself-trainingacrossafricadataset
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
read the original abstract

Identifying the locations and footprints of buildings is vital for many practical and scientific purposes. Such information can be particularly useful in developing regions where alternative data sources may be scarce. In this work, we describe a model training pipeline for detecting buildings across the entire continent of Africa, using 50 cm satellite imagery. Starting with the U-Net model, widely used in satellite image analysis, we study variations in architecture, loss functions, regularization, pre-training, self-training and post-processing that increase instance segmentation performance. Experiments were carried out using a dataset of 100k satellite images across Africa containing 1.75M manually labelled building instances, and further datasets for pre-training and self-training. We report novel methods for improving performance of building detection with this type of model, including the use of mixup (mAP +0.12) and self-training with soft KL loss (mAP +0.06). The resulting pipeline obtains good results even on a wide variety of challenging rural and urban contexts, and was used to create the Open Buildings dataset of 516M Africa-wide detected footprints.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 9 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Toward Seasonal Guidelines for Robust Deep-Learning Sentinel-2 Building Detection in Different Area Types

    cs.CV 2026-07 conditional novelty 6.0 of 10

    Summer Sentinel-2 imagery and a U-Net give the most reliable building detection; winter scenes and low-density settlement types produce large accuracy drops.

  2. Global Building Area Estimation Products: How Accurate Are They?

    cs.CV 2026-07 conditional novelty 6.0 of 10

    An independent multi-grid benchmark finds GBA and TEMPO most accurate global building-area products overall, with accuracy dropping sharply in Africa, Asia, and high-density urban areas.

  3. DeepC4: Deep Conditional Census-Constrained Clustering for Large-scale Multitask Spatial Disaggregation of Urban Morphology

    cs.LG 2025-07 reject novelty 6.0 of 10

    DeepC4 combines multitask deep clustering with sector-level census constraints to map Rwandan building materials and heights from 10 m satellite imagery.

  4. Signal from Space: Detecting Schools and Towers to Bridge the Digital Divide

    cs.CV 2026-07 conditional novelty 5.0 of 10

    Transfer-learned YOLO detectors find schools and towers in Lesotho satellite imagery; CPP line-of-sight within 1 km triages schools as high-potential or hard-to-connect.

  5. Beyond Pretty Pictures: Combined Single- and Multi-Image Super-resolution for Sentinel-2 Images

    eess.IV 2025-05 conditional novelty 5.0 of 10

    SEN4X, a hybrid single- and multi-image super-resolution network, lifts Sentinel-2 imagery to 2.5 m and improves land-cover classification accuracy in Hanoi over SISR, MISR, and stacked-input baselines.

  6. Large-scale School Mapping using Weakly Supervised Deep Learning for Universal School Connectivity

    cs.CV 2024-12 conditional novelty 5.0 of 10

    A weakly supervised satellite-imagery pipeline locates schools in ten African countries with AUPRC above 0.96 and generates nationwide candidate school maps.

  7. Modeling for the Growth of Unorganized Retailing in the Presence of Organized and E-Retailing in Indian Pharmaceutical Industry

    econ.GN 2025-07 reject novelty 4.0 of 10

    An agent-based model calibrated with a 138-person survey suggests that small Indian pharmacies should not push discounts past a tipping point, but this result is built into the model's profit rule.

  8. Life Style Levels: Neighborhood Delineation using Geospatial Data

    cs.CL 2026-07 conditional novelty 3.0 of 10

    A rule-based grid score (median building area / building count) from open building footprints classifies Indian urban neighborhoods into affluence tiers that correlate with loan delinquency and visually match informal...

  9. Comparison of Segmentation Methods in Remote Sensing for Land Use Land Cover

    cs.CV 2025-07 conditional novelty 3.0 of 10

    On Cartosat-3 imagery of Hyderabad, a Cross-Pseudo-Supervision model with dynamic weighting achieves the best MIoU (42.14) while plain CPS attains the best recall (79.59); the analysis documents tree loss and building...

Pith tools