{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:QQUOMX4CFTY4EUCEONJPQHRBES","short_pith_number":"pith:QQUOMX4C","schema_version":"1.0","canonical_sha256":"8428e65f822cf1c250447352f81e2124b4b1c084e8d7cb372615729ed8e0a85d","source":{"kind":"arxiv","id":"2110.03786","version":2},"attestation_state":"computed","paper":{"title":"Efficient large-scale image retrieval with deep feature orthogonality and Hybrid-Swin-Transformers","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Christof Henkel","submitted_at":"2021-10-07T20:41:13Z","abstract_excerpt":"We present an efficient end-to-end pipeline for largescale landmark recognition and retrieval. We show how to combine and enhance concepts from recent research in image retrieval and introduce two architectures especially suited for large-scale landmark identification. A model with deep orthogonal fusion of local and global features (DOLG) using an EfficientNet backbone as well as a novel Hybrid-Swin-Transformer is discussed and details how to train both architectures efficiently using a step-wise approach and a sub-center arcface loss with dynamic margins are provided. Furthermore, we elabora"},"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":"2110.03786","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2021-10-07T20:41:13Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"5620083def3bd259dfe23f87ebf3040974c80c2ba7343d934fe9cba2088d1b6c","abstract_canon_sha256":"d76f1d1ce97ed66d042eec587edeee8fbe4ceb382d2238f0154054f1fb52f09a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:26:20.962520Z","signature_b64":"i6uRt1ICr1rNCPFkh+pIMO71D+wPWdSIOiCR/oDndo0Ouy5OTaJFERT43kxVphMxn2iYDDWxJkgdyLhn7nVjBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8428e65f822cf1c250447352f81e2124b4b1c084e8d7cb372615729ed8e0a85d","last_reissued_at":"2026-07-05T03:26:20.961992Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:26:20.961992Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Efficient large-scale image retrieval with deep feature orthogonality and Hybrid-Swin-Transformers","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Christof Henkel","submitted_at":"2021-10-07T20:41:13Z","abstract_excerpt":"We present an efficient end-to-end pipeline for largescale landmark recognition and retrieval. We show how to combine and enhance concepts from recent research in image retrieval and introduce two architectures especially suited for large-scale landmark identification. A model with deep orthogonal fusion of local and global features (DOLG) using an EfficientNet backbone as well as a novel Hybrid-Swin-Transformer is discussed and details how to train both architectures efficiently using a step-wise approach and a sub-center arcface loss with dynamic margins are provided. Furthermore, we elabora"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2110.03786","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/2110.03786/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":"2110.03786","created_at":"2026-07-05T03:26:20.962048+00:00"},{"alias_kind":"arxiv_version","alias_value":"2110.03786v2","created_at":"2026-07-05T03:26:20.962048+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2110.03786","created_at":"2026-07-05T03:26:20.962048+00:00"},{"alias_kind":"pith_short_12","alias_value":"QQUOMX4CFTY4","created_at":"2026-07-05T03:26:20.962048+00:00"},{"alias_kind":"pith_short_16","alias_value":"QQUOMX4CFTY4EUCE","created_at":"2026-07-05T03:26:20.962048+00:00"},{"alias_kind":"pith_short_8","alias_value":"QQUOMX4C","created_at":"2026-07-05T03:26:20.962048+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QQUOMX4CFTY4EUCEONJPQHRBES","json":"https://pith.science/pith/QQUOMX4CFTY4EUCEONJPQHRBES.json","graph_json":"https://pith.science/api/pith-number/QQUOMX4CFTY4EUCEONJPQHRBES/graph.json","events_json":"https://pith.science/api/pith-number/QQUOMX4CFTY4EUCEONJPQHRBES/events.json","paper":"https://pith.science/paper/QQUOMX4C"},"agent_actions":{"view_html":"https://pith.science/pith/QQUOMX4CFTY4EUCEONJPQHRBES","download_json":"https://pith.science/pith/QQUOMX4CFTY4EUCEONJPQHRBES.json","view_paper":"https://pith.science/paper/QQUOMX4C","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2110.03786&json=true","fetch_graph":"https://pith.science/api/pith-number/QQUOMX4CFTY4EUCEONJPQHRBES/graph.json","fetch_events":"https://pith.science/api/pith-number/QQUOMX4CFTY4EUCEONJPQHRBES/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QQUOMX4CFTY4EUCEONJPQHRBES/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QQUOMX4CFTY4EUCEONJPQHRBES/action/storage_attestation","attest_author":"https://pith.science/pith/QQUOMX4CFTY4EUCEONJPQHRBES/action/author_attestation","sign_citation":"https://pith.science/pith/QQUOMX4CFTY4EUCEONJPQHRBES/action/citation_signature","submit_replication":"https://pith.science/pith/QQUOMX4CFTY4EUCEONJPQHRBES/action/replication_record"}},"created_at":"2026-07-05T03:26:20.962048+00:00","updated_at":"2026-07-05T03:26:20.962048+00:00"}