{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:V7I2ZMEPAUOQ74F5RL5SES4O3E","short_pith_number":"pith:V7I2ZMEP","schema_version":"1.0","canonical_sha256":"afd1acb08f051d0ff0bd8afb224b8ed91167878baad164cfa4d0ea614fd8736b","source":{"kind":"arxiv","id":"2106.02523","version":1},"attestation_state":"computed","paper":{"title":"Hallucination In Object Detection -- A Study In Visual Part Verification","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Bart Vredebregt, Jan C. van Gemert, Osman Semih Kayhan","submitted_at":"2021-06-04T14:47:11Z","abstract_excerpt":"We show that object detectors can hallucinate and detect missing objects; potentially even accurately localized at their expected, but non-existing, position. This is particularly problematic for applications that rely on visual part verification: detecting if an object part is present or absent. We show how popular object detectors hallucinate objects in a visual part verification task and introduce the first visual part verification dataset: DelftBikes, which has 10,000 bike photographs, with 22 densely annotated parts per image, where some parts may be missing. We explicitly annotated an ex"},"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":"2106.02523","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2021-06-04T14:47:11Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"769595c76e88d2176aaebf1750565caf05f3978705093826ee49a9a88c7f34d2","abstract_canon_sha256":"96645fc173374aa62ef213a14f260b84e34ce47d7fa65863b09f43626e07828c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:46:20.204245Z","signature_b64":"q5I8mikGD6jnT/cdCrDZ8pslyACw5mWW1fTBz+WTZq/nEVwoy7Q0O8Zjb6tygsKKmsIQ1KSm61yR1SncLB9vAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"afd1acb08f051d0ff0bd8afb224b8ed91167878baad164cfa4d0ea614fd8736b","last_reissued_at":"2026-07-05T02:46:20.203649Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:46:20.203649Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Hallucination In Object Detection -- A Study In Visual Part Verification","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Bart Vredebregt, Jan C. van Gemert, Osman Semih Kayhan","submitted_at":"2021-06-04T14:47:11Z","abstract_excerpt":"We show that object detectors can hallucinate and detect missing objects; potentially even accurately localized at their expected, but non-existing, position. This is particularly problematic for applications that rely on visual part verification: detecting if an object part is present or absent. We show how popular object detectors hallucinate objects in a visual part verification task and introduce the first visual part verification dataset: DelftBikes, which has 10,000 bike photographs, with 22 densely annotated parts per image, where some parts may be missing. We explicitly annotated an ex"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2106.02523","kind":"arxiv","version":1},"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/2106.02523/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":"2106.02523","created_at":"2026-07-05T02:46:20.203766+00:00"},{"alias_kind":"arxiv_version","alias_value":"2106.02523v1","created_at":"2026-07-05T02:46:20.203766+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2106.02523","created_at":"2026-07-05T02:46:20.203766+00:00"},{"alias_kind":"pith_short_12","alias_value":"V7I2ZMEPAUOQ","created_at":"2026-07-05T02:46:20.203766+00:00"},{"alias_kind":"pith_short_16","alias_value":"V7I2ZMEPAUOQ74F5","created_at":"2026-07-05T02:46:20.203766+00:00"},{"alias_kind":"pith_short_8","alias_value":"V7I2ZMEP","created_at":"2026-07-05T02:46:20.203766+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.08612","citing_title":"Data-Efficient Challenges in Visual Inductive Priors: A Retrospective","ref_index":78,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/V7I2ZMEPAUOQ74F5RL5SES4O3E","json":"https://pith.science/pith/V7I2ZMEPAUOQ74F5RL5SES4O3E.json","graph_json":"https://pith.science/api/pith-number/V7I2ZMEPAUOQ74F5RL5SES4O3E/graph.json","events_json":"https://pith.science/api/pith-number/V7I2ZMEPAUOQ74F5RL5SES4O3E/events.json","paper":"https://pith.science/paper/V7I2ZMEP"},"agent_actions":{"view_html":"https://pith.science/pith/V7I2ZMEPAUOQ74F5RL5SES4O3E","download_json":"https://pith.science/pith/V7I2ZMEPAUOQ74F5RL5SES4O3E.json","view_paper":"https://pith.science/paper/V7I2ZMEP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2106.02523&json=true","fetch_graph":"https://pith.science/api/pith-number/V7I2ZMEPAUOQ74F5RL5SES4O3E/graph.json","fetch_events":"https://pith.science/api/pith-number/V7I2ZMEPAUOQ74F5RL5SES4O3E/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/V7I2ZMEPAUOQ74F5RL5SES4O3E/action/timestamp_anchor","attest_storage":"https://pith.science/pith/V7I2ZMEPAUOQ74F5RL5SES4O3E/action/storage_attestation","attest_author":"https://pith.science/pith/V7I2ZMEPAUOQ74F5RL5SES4O3E/action/author_attestation","sign_citation":"https://pith.science/pith/V7I2ZMEPAUOQ74F5RL5SES4O3E/action/citation_signature","submit_replication":"https://pith.science/pith/V7I2ZMEPAUOQ74F5RL5SES4O3E/action/replication_record"}},"created_at":"2026-07-05T02:46:20.203766+00:00","updated_at":"2026-07-05T02:46:20.203766+00:00"}