{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:JIKD4656WZSV4O27UGA3XTWEPS","short_pith_number":"pith:JIKD4656","schema_version":"1.0","canonical_sha256":"4a143e7bbeb6655e3b5fa181bbcec47ca1c1da67436ea36c7fbde83b11f41335","source":{"kind":"arxiv","id":"2410.16268","version":3},"attestation_state":"computed","paper":{"title":"SAM2Long: Enhancing SAM 2 for Long Video Segmentation with a Training-Free Memory Tree","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Dahua Lin, Jiaqi Wang, Pan Zhang, Rui Qian, Shuangrui Ding, Xiaoyi Dong, Yuhang Cao, Yuhang Zang, Yuwei Guo","submitted_at":"2024-10-21T17:59:19Z","abstract_excerpt":"The Segment Anything Model 2 (SAM 2) has emerged as a powerful foundation model for object segmentation in both images and videos, paving the way for various downstream video applications. The crucial design of SAM 2 for video segmentation is its memory module, which prompts object-aware memories from previous frames for current frame prediction. However, its greedy-selection memory design suffers from the \"error accumulation\" problem, where an errored or missed mask will cascade and influence the segmentation of the subsequent frames, which limits the performance of SAM 2 toward complex long-"},"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":"2410.16268","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-10-21T17:59:19Z","cross_cats_sorted":[],"title_canon_sha256":"d757edc56e1114f14de42c47c63e4bd42f3b374f58142f3b281a926ed38fcf88","abstract_canon_sha256":"c6b810a991353482caa73fe34a6108e836b59658227c7cbc64ae2899d04891b8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:44:42.317888Z","signature_b64":"8Nvo7xEgLVSpmRN29mU95ajPwd88P+zhE2+ir6WZfgop14IXo1y0usWlQRQyJkj3dZBCReNKY7hx7VTp85uLDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4a143e7bbeb6655e3b5fa181bbcec47ca1c1da67436ea36c7fbde83b11f41335","last_reissued_at":"2026-07-05T11:44:42.317404Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:44:42.317404Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SAM2Long: Enhancing SAM 2 for Long Video Segmentation with a Training-Free Memory Tree","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Dahua Lin, Jiaqi Wang, Pan Zhang, Rui Qian, Shuangrui Ding, Xiaoyi Dong, Yuhang Cao, Yuhang Zang, Yuwei Guo","submitted_at":"2024-10-21T17:59:19Z","abstract_excerpt":"The Segment Anything Model 2 (SAM 2) has emerged as a powerful foundation model for object segmentation in both images and videos, paving the way for various downstream video applications. The crucial design of SAM 2 for video segmentation is its memory module, which prompts object-aware memories from previous frames for current frame prediction. However, its greedy-selection memory design suffers from the \"error accumulation\" problem, where an errored or missed mask will cascade and influence the segmentation of the subsequent frames, which limits the performance of SAM 2 toward complex long-"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.16268","kind":"arxiv","version":3},"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/2410.16268/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":"2410.16268","created_at":"2026-07-05T11:44:42.317464+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.16268v3","created_at":"2026-07-05T11:44:42.317464+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.16268","created_at":"2026-07-05T11:44:42.317464+00:00"},{"alias_kind":"pith_short_12","alias_value":"JIKD4656WZSV","created_at":"2026-07-05T11:44:42.317464+00:00"},{"alias_kind":"pith_short_16","alias_value":"JIKD4656WZSV4O27","created_at":"2026-07-05T11:44:42.317464+00:00"},{"alias_kind":"pith_short_8","alias_value":"JIKD4656","created_at":"2026-07-05T11:44:42.317464+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":7,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.24449","citing_title":"SENTRY: SAM2-Enhanced Neighbor-Aware and Temporally Reasoned Memory for Visual Tracking","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2510.18822","citing_title":"SAM 2++: Tracking Anything at Any Granularity","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2511.16719","citing_title":"SAM 3: Segment Anything with Concepts","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2512.22046","citing_title":"Backdoor Attacks on Prompt-Driven Video Segmentation Foundation Models","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2601.08831","citing_title":"3AM: 3egment Anything with Geometric Consistency in Videos","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2603.08096","citing_title":"TrianguLang: Geometry-Aware Semantic Consensus for Pose-Free 3D Localization","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2604.22162","citing_title":"SAMIDARE: Advanced Tracking-by-Segmentation for Dense Scenarios","ref_index":5,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JIKD4656WZSV4O27UGA3XTWEPS","json":"https://pith.science/pith/JIKD4656WZSV4O27UGA3XTWEPS.json","graph_json":"https://pith.science/api/pith-number/JIKD4656WZSV4O27UGA3XTWEPS/graph.json","events_json":"https://pith.science/api/pith-number/JIKD4656WZSV4O27UGA3XTWEPS/events.json","paper":"https://pith.science/paper/JIKD4656"},"agent_actions":{"view_html":"https://pith.science/pith/JIKD4656WZSV4O27UGA3XTWEPS","download_json":"https://pith.science/pith/JIKD4656WZSV4O27UGA3XTWEPS.json","view_paper":"https://pith.science/paper/JIKD4656","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.16268&json=true","fetch_graph":"https://pith.science/api/pith-number/JIKD4656WZSV4O27UGA3XTWEPS/graph.json","fetch_events":"https://pith.science/api/pith-number/JIKD4656WZSV4O27UGA3XTWEPS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JIKD4656WZSV4O27UGA3XTWEPS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JIKD4656WZSV4O27UGA3XTWEPS/action/storage_attestation","attest_author":"https://pith.science/pith/JIKD4656WZSV4O27UGA3XTWEPS/action/author_attestation","sign_citation":"https://pith.science/pith/JIKD4656WZSV4O27UGA3XTWEPS/action/citation_signature","submit_replication":"https://pith.science/pith/JIKD4656WZSV4O27UGA3XTWEPS/action/replication_record"}},"created_at":"2026-07-05T11:44:42.317464+00:00","updated_at":"2026-07-05T11:44:42.317464+00:00"}