{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:2YM33OLDZGIMD3VFE7SUQFUVCJ","short_pith_number":"pith:2YM33OLD","schema_version":"1.0","canonical_sha256":"d619bdb963c990c1eea527e548169512508d2b5214a2b66eb2323bbb1465c155","source":{"kind":"arxiv","id":"2311.18512","version":2},"attestation_state":"computed","paper":{"title":"Union-over-Intersections: Object Detection beyond Winner-Takes-All","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Aritra Bhowmik, Cees G. M. Snoek, Martin R. Oswald, Pascal Mettes","submitted_at":"2023-11-30T12:40:23Z","abstract_excerpt":"This paper revisits the problem of predicting box locations in object detection architectures. Typically, each box proposal or box query aims to directly maximize the intersection-over-union score with the ground truth, followed by a winner-takes-all non-maximum suppression where only the highest scoring box in each region is retained. We observe that both steps are sub-optimal: the first involves regressing proposals to the entire ground truth, which is a difficult task even with large receptive fields, and the second neglects valuable information from boxes other than the top candidate. Inst"},"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":"2311.18512","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-11-30T12:40:23Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"8cb5b44edd1bb6e063774549e39a3ea21a365ef06f5752e4b6103954c421ee95","abstract_canon_sha256":"027a8652ae9c68fda10e8c9a4626cfa3e941cc945e86b37df351dce9244d1092"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:51:29.890262Z","signature_b64":"OhEkWrLloy8BdLoKzjNjmS0QN280psIbFj9zvN1ej+YXYaBUgZffkeTij90oIzbT64lmaeiMyZVLiM1drQmaBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d619bdb963c990c1eea527e548169512508d2b5214a2b66eb2323bbb1465c155","last_reissued_at":"2026-07-05T09:51:29.889814Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:51:29.889814Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Union-over-Intersections: Object Detection beyond Winner-Takes-All","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Aritra Bhowmik, Cees G. M. Snoek, Martin R. Oswald, Pascal Mettes","submitted_at":"2023-11-30T12:40:23Z","abstract_excerpt":"This paper revisits the problem of predicting box locations in object detection architectures. Typically, each box proposal or box query aims to directly maximize the intersection-over-union score with the ground truth, followed by a winner-takes-all non-maximum suppression where only the highest scoring box in each region is retained. We observe that both steps are sub-optimal: the first involves regressing proposals to the entire ground truth, which is a difficult task even with large receptive fields, and the second neglects valuable information from boxes other than the top candidate. Inst"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.18512","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/2311.18512/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":"2311.18512","created_at":"2026-07-05T09:51:29.889873+00:00"},{"alias_kind":"arxiv_version","alias_value":"2311.18512v2","created_at":"2026-07-05T09:51:29.889873+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.18512","created_at":"2026-07-05T09:51:29.889873+00:00"},{"alias_kind":"pith_short_12","alias_value":"2YM33OLDZGIM","created_at":"2026-07-05T09:51:29.889873+00:00"},{"alias_kind":"pith_short_16","alias_value":"2YM33OLDZGIMD3VF","created_at":"2026-07-05T09:51:29.889873+00:00"},{"alias_kind":"pith_short_8","alias_value":"2YM33OLD","created_at":"2026-07-05T09:51:29.889873+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.13457","citing_title":"Deep Learning-Based Multi-Object Tracking: A Comprehensive Survey from Foundations to State-of-the-Art","ref_index":9,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2YM33OLDZGIMD3VFE7SUQFUVCJ","json":"https://pith.science/pith/2YM33OLDZGIMD3VFE7SUQFUVCJ.json","graph_json":"https://pith.science/api/pith-number/2YM33OLDZGIMD3VFE7SUQFUVCJ/graph.json","events_json":"https://pith.science/api/pith-number/2YM33OLDZGIMD3VFE7SUQFUVCJ/events.json","paper":"https://pith.science/paper/2YM33OLD"},"agent_actions":{"view_html":"https://pith.science/pith/2YM33OLDZGIMD3VFE7SUQFUVCJ","download_json":"https://pith.science/pith/2YM33OLDZGIMD3VFE7SUQFUVCJ.json","view_paper":"https://pith.science/paper/2YM33OLD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2311.18512&json=true","fetch_graph":"https://pith.science/api/pith-number/2YM33OLDZGIMD3VFE7SUQFUVCJ/graph.json","fetch_events":"https://pith.science/api/pith-number/2YM33OLDZGIMD3VFE7SUQFUVCJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2YM33OLDZGIMD3VFE7SUQFUVCJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2YM33OLDZGIMD3VFE7SUQFUVCJ/action/storage_attestation","attest_author":"https://pith.science/pith/2YM33OLDZGIMD3VFE7SUQFUVCJ/action/author_attestation","sign_citation":"https://pith.science/pith/2YM33OLDZGIMD3VFE7SUQFUVCJ/action/citation_signature","submit_replication":"https://pith.science/pith/2YM33OLDZGIMD3VFE7SUQFUVCJ/action/replication_record"}},"created_at":"2026-07-05T09:51:29.889873+00:00","updated_at":"2026-07-05T09:51:29.889873+00:00"}