{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2017:3FSSMXMGX4ZFOIRPDETS4ZP5BU","short_pith_number":"pith:3FSSMXMG","schema_version":"1.0","canonical_sha256":"d965265d86bf3257222f19272e65fd0d3c0f961984e446fefaebf16738df53e0","source":{"kind":"arxiv","id":"1711.00804","version":2},"attestation_state":"computed","paper":{"title":"Framework for evaluation of sound event detection in web videos","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.IR","eess.AS"],"primary_cat":"cs.SD","authors_text":"Ankit Shah, Anurag Kumar, Benjamin Elizalde, Bhiksha Raj, Rohan Badlani","submitted_at":"2017-11-02T16:32:23Z","abstract_excerpt":"The largest source of sound events is web videos. Most videos lack sound event labels at segment level, however, a significant number of them do respond to text queries, from a match found using metadata by search engines. In this paper we explore the extent to which a search query can be used as the true label for detection of sound events in videos. We present a framework for large-scale sound event recognition on web videos. The framework crawls videos using search queries corresponding to 78 sound event labels drawn from three datasets. The datasets are used to train three classifiers, and"},"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":"1711.00804","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.SD","submitted_at":"2017-11-02T16:32:23Z","cross_cats_sorted":["cs.AI","cs.IR","eess.AS"],"title_canon_sha256":"c4ad4808dd52768a3084502c0e0f0944705131684d435f689b5a38e7a71241b0","abstract_canon_sha256":"754b0457dfa656e501eb24bfe2a98d7230b8069a9407fd71979e5f6ae1993657"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:19:16.889999Z","signature_b64":"PGVkvZ1KOx83jV0EF/wyMLMVo1qIHENwp3ellUXeoyuMnOoasb6CDFEPdfNloT6K+OgE0ztmckaRhkwmCNecDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d965265d86bf3257222f19272e65fd0d3c0f961984e446fefaebf16738df53e0","last_reissued_at":"2026-05-18T00:19:16.889444Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:19:16.889444Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Framework for evaluation of sound event detection in web videos","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.IR","eess.AS"],"primary_cat":"cs.SD","authors_text":"Ankit Shah, Anurag Kumar, Benjamin Elizalde, Bhiksha Raj, Rohan Badlani","submitted_at":"2017-11-02T16:32:23Z","abstract_excerpt":"The largest source of sound events is web videos. Most videos lack sound event labels at segment level, however, a significant number of them do respond to text queries, from a match found using metadata by search engines. In this paper we explore the extent to which a search query can be used as the true label for detection of sound events in videos. We present a framework for large-scale sound event recognition on web videos. The framework crawls videos using search queries corresponding to 78 sound event labels drawn from three datasets. The datasets are used to train three classifiers, and"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1711.00804","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":"1711.00804","created_at":"2026-05-18T00:19:16.889550+00:00"},{"alias_kind":"arxiv_version","alias_value":"1711.00804v2","created_at":"2026-05-18T00:19:16.889550+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1711.00804","created_at":"2026-05-18T00:19:16.889550+00:00"},{"alias_kind":"pith_short_12","alias_value":"3FSSMXMGX4ZF","created_at":"2026-05-18T12:30:58.224056+00:00"},{"alias_kind":"pith_short_16","alias_value":"3FSSMXMGX4ZFOIRP","created_at":"2026-05-18T12:30:58.224056+00:00"},{"alias_kind":"pith_short_8","alias_value":"3FSSMXMG","created_at":"2026-05-18T12:30:58.224056+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/3FSSMXMGX4ZFOIRPDETS4ZP5BU","json":"https://pith.science/pith/3FSSMXMGX4ZFOIRPDETS4ZP5BU.json","graph_json":"https://pith.science/api/pith-number/3FSSMXMGX4ZFOIRPDETS4ZP5BU/graph.json","events_json":"https://pith.science/api/pith-number/3FSSMXMGX4ZFOIRPDETS4ZP5BU/events.json","paper":"https://pith.science/paper/3FSSMXMG"},"agent_actions":{"view_html":"https://pith.science/pith/3FSSMXMGX4ZFOIRPDETS4ZP5BU","download_json":"https://pith.science/pith/3FSSMXMGX4ZFOIRPDETS4ZP5BU.json","view_paper":"https://pith.science/paper/3FSSMXMG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1711.00804&json=true","fetch_graph":"https://pith.science/api/pith-number/3FSSMXMGX4ZFOIRPDETS4ZP5BU/graph.json","fetch_events":"https://pith.science/api/pith-number/3FSSMXMGX4ZFOIRPDETS4ZP5BU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3FSSMXMGX4ZFOIRPDETS4ZP5BU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3FSSMXMGX4ZFOIRPDETS4ZP5BU/action/storage_attestation","attest_author":"https://pith.science/pith/3FSSMXMGX4ZFOIRPDETS4ZP5BU/action/author_attestation","sign_citation":"https://pith.science/pith/3FSSMXMGX4ZFOIRPDETS4ZP5BU/action/citation_signature","submit_replication":"https://pith.science/pith/3FSSMXMGX4ZFOIRPDETS4ZP5BU/action/replication_record"}},"created_at":"2026-05-18T00:19:16.889550+00:00","updated_at":"2026-05-18T00:19:16.889550+00:00"}