{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:7YFGFZ74VK3T5XYKD2RV2HQIB2","short_pith_number":"pith:7YFGFZ74","schema_version":"1.0","canonical_sha256":"fe0a62e7fcaab73edf0a1ea35d1e080ea79f5e5c22d098feab6c487ccf3384e7","source":{"kind":"arxiv","id":"2107.12283","version":2},"attestation_state":"computed","paper":{"title":"Continental-Scale Building Detection from High Resolution Satellite Imagery","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Abigail Annkah, Daniel Keysers, John Quinn, Marvin Ritter, Maxim Neumann, Moustapha Cisse, Sergii Kashubin, Wojciech Sirko, Yann Dauphin, Yasser Salah Eddine Bouchareb","submitted_at":"2021-07-26T15:48:14Z","abstract_excerpt":"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 performanc"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2107.12283","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2021-07-26T15:48:14Z","cross_cats_sorted":[],"title_canon_sha256":"fb5deba89520ab2df87c53b498884d2f070ef034b64044ec1bf816cf2a6c0ede","abstract_canon_sha256":"6158378a1ca9613feaa6a2b63f000616c17c77045784b6c8921c16dfaa8f7ed7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:01:57.189012Z","signature_b64":"IiP3hY40E/OZAorDuJUB+nP238HZPkKlJHKJ9oUSZJqRwBMqaCZnT+aCEKUf/6p0NgxNKoYIAucAA+IlaRIXBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fe0a62e7fcaab73edf0a1ea35d1e080ea79f5e5c22d098feab6c487ccf3384e7","last_reissued_at":"2026-07-05T03:01:57.188568Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:01:57.188568Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Continental-Scale Building Detection from High Resolution Satellite Imagery","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Abigail Annkah, Daniel Keysers, John Quinn, Marvin Ritter, Maxim Neumann, Moustapha Cisse, Sergii Kashubin, Wojciech Sirko, Yann Dauphin, Yasser Salah Eddine Bouchareb","submitted_at":"2021-07-26T15:48:14Z","abstract_excerpt":"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 performanc"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2107.12283","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2107.12283/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2107.12283","created_at":"2026-07-05T03:01:57.188626+00:00"},{"alias_kind":"arxiv_version","alias_value":"2107.12283v2","created_at":"2026-07-05T03:01:57.188626+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2107.12283","created_at":"2026-07-05T03:01:57.188626+00:00"},{"alias_kind":"pith_short_12","alias_value":"7YFGFZ74VK3T","created_at":"2026-07-05T03:01:57.188626+00:00"},{"alias_kind":"pith_short_16","alias_value":"7YFGFZ74VK3T5XYK","created_at":"2026-07-05T03:01:57.188626+00:00"},{"alias_kind":"pith_short_8","alias_value":"7YFGFZ74","created_at":"2026-07-05T03:01:57.188626+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.06529","citing_title":"Life Style Levels: Neighborhood Delineation using Geospatial Data","ref_index":7,"is_internal_anchor":true},{"citing_arxiv_id":"2605.03968","citing_title":"Label-Efficient School Detection from Aerial Imagery via Weakly Supervised Pretraining and Fine-Tuning","ref_index":29,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7YFGFZ74VK3T5XYKD2RV2HQIB2","json":"https://pith.science/pith/7YFGFZ74VK3T5XYKD2RV2HQIB2.json","graph_json":"https://pith.science/api/pith-number/7YFGFZ74VK3T5XYKD2RV2HQIB2/graph.json","events_json":"https://pith.science/api/pith-number/7YFGFZ74VK3T5XYKD2RV2HQIB2/events.json","paper":"https://pith.science/paper/7YFGFZ74"},"agent_actions":{"view_html":"https://pith.science/pith/7YFGFZ74VK3T5XYKD2RV2HQIB2","download_json":"https://pith.science/pith/7YFGFZ74VK3T5XYKD2RV2HQIB2.json","view_paper":"https://pith.science/paper/7YFGFZ74","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2107.12283&json=true","fetch_graph":"https://pith.science/api/pith-number/7YFGFZ74VK3T5XYKD2RV2HQIB2/graph.json","fetch_events":"https://pith.science/api/pith-number/7YFGFZ74VK3T5XYKD2RV2HQIB2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7YFGFZ74VK3T5XYKD2RV2HQIB2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7YFGFZ74VK3T5XYKD2RV2HQIB2/action/storage_attestation","attest_author":"https://pith.science/pith/7YFGFZ74VK3T5XYKD2RV2HQIB2/action/author_attestation","sign_citation":"https://pith.science/pith/7YFGFZ74VK3T5XYKD2RV2HQIB2/action/citation_signature","submit_replication":"https://pith.science/pith/7YFGFZ74VK3T5XYKD2RV2HQIB2/action/replication_record"}},"created_at":"2026-07-05T03:01:57.188626+00:00","updated_at":"2026-07-05T03:01:57.188626+00:00"}