{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2016:FSIBM4L2FH5AU4WCVGYX46ZLAJ","short_pith_number":"pith:FSIBM4L2","schema_version":"1.0","canonical_sha256":"2c9016717a29fa0a72c2a9b17e7b2b027b8a74940c773f2e8b382c4752ce3a99","source":{"kind":"arxiv","id":"1607.03476","version":2},"attestation_state":"computed","paper":{"title":"End-to-end training of object class detectors for mean average precision","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Paul Henderson, Vittorio Ferrari","submitted_at":"2016-07-12T19:45:12Z","abstract_excerpt":"We present a method for training CNN-based object class detectors directly using mean average precision (mAP) as the training loss, in a truly end-to-end fashion that includes non-maximum suppression (NMS) at training time. This contrasts with the traditional approach of training a CNN for a window classification loss, then applying NMS only at test time, when mAP is used as the evaluation metric in place of classification accuracy. However, mAP following NMS forms a piecewise-constant structured loss over thousands of windows, with gradients that do not convey useful information for gradient "},"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":"1607.03476","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2016-07-12T19:45:12Z","cross_cats_sorted":[],"title_canon_sha256":"92c4baee1c283db9fb937db8ba3f1df76568f1d45c71e2566a1d3b9982695aee","abstract_canon_sha256":"ce88ac07c9063c507b8e56dc0f265f7cdbfb58cce88d5da14c4694da17af15f2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:48:37.231035Z","signature_b64":"7bXy5waoZDtCGhDVlxex7GELiUe3QYegyxtCOKSnyBmCMrKeR2NcK4wkV0n1q9S4MswvvEHZWKilRCkY7H9ZAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2c9016717a29fa0a72c2a9b17e7b2b027b8a74940c773f2e8b382c4752ce3a99","last_reissued_at":"2026-05-18T00:48:37.230352Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:48:37.230352Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"End-to-end training of object class detectors for mean average precision","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Paul Henderson, Vittorio Ferrari","submitted_at":"2016-07-12T19:45:12Z","abstract_excerpt":"We present a method for training CNN-based object class detectors directly using mean average precision (mAP) as the training loss, in a truly end-to-end fashion that includes non-maximum suppression (NMS) at training time. This contrasts with the traditional approach of training a CNN for a window classification loss, then applying NMS only at test time, when mAP is used as the evaluation metric in place of classification accuracy. However, mAP following NMS forms a piecewise-constant structured loss over thousands of windows, with gradients that do not convey useful information for gradient "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1607.03476","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":""},"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":"1607.03476","created_at":"2026-05-18T00:48:37.230474+00:00"},{"alias_kind":"arxiv_version","alias_value":"1607.03476v2","created_at":"2026-05-18T00:48:37.230474+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1607.03476","created_at":"2026-05-18T00:48:37.230474+00:00"},{"alias_kind":"pith_short_12","alias_value":"FSIBM4L2FH5A","created_at":"2026-05-18T12:30:15.759754+00:00"},{"alias_kind":"pith_short_16","alias_value":"FSIBM4L2FH5AU4WC","created_at":"2026-05-18T12:30:15.759754+00:00"},{"alias_kind":"pith_short_8","alias_value":"FSIBM4L2","created_at":"2026-05-18T12:30:15.759754+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/FSIBM4L2FH5AU4WCVGYX46ZLAJ","json":"https://pith.science/pith/FSIBM4L2FH5AU4WCVGYX46ZLAJ.json","graph_json":"https://pith.science/api/pith-number/FSIBM4L2FH5AU4WCVGYX46ZLAJ/graph.json","events_json":"https://pith.science/api/pith-number/FSIBM4L2FH5AU4WCVGYX46ZLAJ/events.json","paper":"https://pith.science/paper/FSIBM4L2"},"agent_actions":{"view_html":"https://pith.science/pith/FSIBM4L2FH5AU4WCVGYX46ZLAJ","download_json":"https://pith.science/pith/FSIBM4L2FH5AU4WCVGYX46ZLAJ.json","view_paper":"https://pith.science/paper/FSIBM4L2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1607.03476&json=true","fetch_graph":"https://pith.science/api/pith-number/FSIBM4L2FH5AU4WCVGYX46ZLAJ/graph.json","fetch_events":"https://pith.science/api/pith-number/FSIBM4L2FH5AU4WCVGYX46ZLAJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FSIBM4L2FH5AU4WCVGYX46ZLAJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FSIBM4L2FH5AU4WCVGYX46ZLAJ/action/storage_attestation","attest_author":"https://pith.science/pith/FSIBM4L2FH5AU4WCVGYX46ZLAJ/action/author_attestation","sign_citation":"https://pith.science/pith/FSIBM4L2FH5AU4WCVGYX46ZLAJ/action/citation_signature","submit_replication":"https://pith.science/pith/FSIBM4L2FH5AU4WCVGYX46ZLAJ/action/replication_record"}},"created_at":"2026-05-18T00:48:37.230474+00:00","updated_at":"2026-05-18T00:48:37.230474+00:00"}