{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:QE6FHIQY37FASLMEAJ6NO3MCPG","short_pith_number":"pith:QE6FHIQY","schema_version":"1.0","canonical_sha256":"813c53a218dfca092d84027cd76d8279b032193c0c2bbbddc644331049996bd0","source":{"kind":"arxiv","id":"2303.09044","version":5},"attestation_state":"computed","paper":{"title":"CoLo-CAM: Class Activation Mapping for Object Co-Localization in Weakly-Labeled Unconstrained Videos","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Eric Granger, Ismail Ben Ayed, Luke McCaffrey, Marco Pedersoli, Shakeeb Murtaza, Soufiane Belharbi","submitted_at":"2023-03-16T02:29:53Z","abstract_excerpt":"Leveraging spatiotemporal information in videos is critical for weakly supervised video object localization (WSVOL) tasks. However, state-of-the-art methods only rely on visual and motion cues, while discarding discriminative information, making them susceptible to inaccurate localizations. Recently, discriminative models have been explored for WSVOL tasks using a temporal class activation mapping (CAM) method. Although their results are promising, objects are assumed to have limited movement from frame to frame, leading to degradation in performance for relatively long-term dependencies. This"},"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":"2303.09044","kind":"arxiv","version":5},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CV","submitted_at":"2023-03-16T02:29:53Z","cross_cats_sorted":[],"title_canon_sha256":"8372fa8439c1371441dbb5c1fbd0bbcf8c9a98d280192509143c286b311423a4","abstract_canon_sha256":"f1d2afc821e929c7e5523cf4b16ab38cd64f3dbb0ed9edde8a2b45104ad828df"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:05:06.573420Z","signature_b64":"aOsDyKfyECOOl4DvpxbSgPz9m4PBAgTL8F0+PKifKYtn8r4uVEpSr2n+eUrO9n7Nw7fWOopRUj5MyGQxYiqfDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"813c53a218dfca092d84027cd76d8279b032193c0c2bbbddc644331049996bd0","last_reissued_at":"2026-07-05T10:05:06.572960Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:05:06.572960Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CoLo-CAM: Class Activation Mapping for Object Co-Localization in Weakly-Labeled Unconstrained Videos","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Eric Granger, Ismail Ben Ayed, Luke McCaffrey, Marco Pedersoli, Shakeeb Murtaza, Soufiane Belharbi","submitted_at":"2023-03-16T02:29:53Z","abstract_excerpt":"Leveraging spatiotemporal information in videos is critical for weakly supervised video object localization (WSVOL) tasks. However, state-of-the-art methods only rely on visual and motion cues, while discarding discriminative information, making them susceptible to inaccurate localizations. Recently, discriminative models have been explored for WSVOL tasks using a temporal class activation mapping (CAM) method. Although their results are promising, objects are assumed to have limited movement from frame to frame, leading to degradation in performance for relatively long-term dependencies. This"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2303.09044","kind":"arxiv","version":5},"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/2303.09044/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":"2303.09044","created_at":"2026-07-05T10:05:06.573016+00:00"},{"alias_kind":"arxiv_version","alias_value":"2303.09044v5","created_at":"2026-07-05T10:05:06.573016+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2303.09044","created_at":"2026-07-05T10:05:06.573016+00:00"},{"alias_kind":"pith_short_12","alias_value":"QE6FHIQY37FA","created_at":"2026-07-05T10:05:06.573016+00:00"},{"alias_kind":"pith_short_16","alias_value":"QE6FHIQY37FASLME","created_at":"2026-07-05T10:05:06.573016+00:00"},{"alias_kind":"pith_short_8","alias_value":"QE6FHIQY","created_at":"2026-07-05T10:05:06.573016+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2509.00752","citing_title":"Multi-Level CLS Token Fusion for Contrastive Learning in Endoscopy Image Classification","ref_index":20,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QE6FHIQY37FASLMEAJ6NO3MCPG","json":"https://pith.science/pith/QE6FHIQY37FASLMEAJ6NO3MCPG.json","graph_json":"https://pith.science/api/pith-number/QE6FHIQY37FASLMEAJ6NO3MCPG/graph.json","events_json":"https://pith.science/api/pith-number/QE6FHIQY37FASLMEAJ6NO3MCPG/events.json","paper":"https://pith.science/paper/QE6FHIQY"},"agent_actions":{"view_html":"https://pith.science/pith/QE6FHIQY37FASLMEAJ6NO3MCPG","download_json":"https://pith.science/pith/QE6FHIQY37FASLMEAJ6NO3MCPG.json","view_paper":"https://pith.science/paper/QE6FHIQY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2303.09044&json=true","fetch_graph":"https://pith.science/api/pith-number/QE6FHIQY37FASLMEAJ6NO3MCPG/graph.json","fetch_events":"https://pith.science/api/pith-number/QE6FHIQY37FASLMEAJ6NO3MCPG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QE6FHIQY37FASLMEAJ6NO3MCPG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QE6FHIQY37FASLMEAJ6NO3MCPG/action/storage_attestation","attest_author":"https://pith.science/pith/QE6FHIQY37FASLMEAJ6NO3MCPG/action/author_attestation","sign_citation":"https://pith.science/pith/QE6FHIQY37FASLMEAJ6NO3MCPG/action/citation_signature","submit_replication":"https://pith.science/pith/QE6FHIQY37FASLMEAJ6NO3MCPG/action/replication_record"}},"created_at":"2026-07-05T10:05:06.573016+00:00","updated_at":"2026-07-05T10:05:06.573016+00:00"}