{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:DDDQXKYHG27GELLBAY472UG6GB","short_pith_number":"pith:DDDQXKYH","schema_version":"1.0","canonical_sha256":"18c70bab0736be622d610639fd50de30427b2a785237a35ad96ba06f92fddd5f","source":{"kind":"arxiv","id":"2501.13667","version":5},"attestation_state":"computed","paper":{"title":"MPG-SAM 2: Adapting SAM 2 with Mask Priors and Global Context for Referring Video Object Segmentation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Fu Rong, Lefei Zhang, Meng Lan, Qian Zhang","submitted_at":"2025-01-23T13:53:33Z","abstract_excerpt":"Referring video object segmentation (RVOS) aims to segment objects in a video according to textual descriptions, which requires the integration of multimodal information and temporal dynamics perception. The Segment Anything Model 2 (SAM 2) has shown great effectiveness across various video segmentation tasks. However, its application to offline RVOS is challenged by the translation of the text into effective prompts and a lack of global context awareness. In this paper, we propose a novel RVOS framework, termed MPG-SAM 2, to address these challenges. Specifically, MPG-SAM 2 employs a unified "},"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":"2501.13667","kind":"arxiv","version":5},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-01-23T13:53:33Z","cross_cats_sorted":[],"title_canon_sha256":"78de7dd2145cc2212c43891af31b9dc6b5d61891ea336dc1ab6ce0c5b3e11926","abstract_canon_sha256":"159a42ad18a4779e02ff36d15847e2b451970efbc6a3ccd403bbeb55768a8fde"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:50:30.269745Z","signature_b64":"EYweUo3vaPlZgeWwqNnzQZ7bIZkKh7En4ewM3HzkjWAXCIeVGjrXHM6q2V5JxPei5Q5mW2atLviWaL2ztsLeAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"18c70bab0736be622d610639fd50de30427b2a785237a35ad96ba06f92fddd5f","last_reissued_at":"2026-07-05T11:50:30.269249Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:50:30.269249Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MPG-SAM 2: Adapting SAM 2 with Mask Priors and Global Context for Referring Video Object Segmentation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Fu Rong, Lefei Zhang, Meng Lan, Qian Zhang","submitted_at":"2025-01-23T13:53:33Z","abstract_excerpt":"Referring video object segmentation (RVOS) aims to segment objects in a video according to textual descriptions, which requires the integration of multimodal information and temporal dynamics perception. The Segment Anything Model 2 (SAM 2) has shown great effectiveness across various video segmentation tasks. However, its application to offline RVOS is challenged by the translation of the text into effective prompts and a lack of global context awareness. In this paper, we propose a novel RVOS framework, termed MPG-SAM 2, to address these challenges. Specifically, MPG-SAM 2 employs a unified "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.13667","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/2501.13667/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":"2501.13667","created_at":"2026-07-05T11:50:30.269305+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.13667v5","created_at":"2026-07-05T11:50:30.269305+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.13667","created_at":"2026-07-05T11:50:30.269305+00:00"},{"alias_kind":"pith_short_12","alias_value":"DDDQXKYHG27G","created_at":"2026-07-05T11:50:30.269305+00:00"},{"alias_kind":"pith_short_16","alias_value":"DDDQXKYHG27GELLB","created_at":"2026-07-05T11:50:30.269305+00:00"},{"alias_kind":"pith_short_8","alias_value":"DDDQXKYH","created_at":"2026-07-05T11:50:30.269305+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/DDDQXKYHG27GELLBAY472UG6GB","json":"https://pith.science/pith/DDDQXKYHG27GELLBAY472UG6GB.json","graph_json":"https://pith.science/api/pith-number/DDDQXKYHG27GELLBAY472UG6GB/graph.json","events_json":"https://pith.science/api/pith-number/DDDQXKYHG27GELLBAY472UG6GB/events.json","paper":"https://pith.science/paper/DDDQXKYH"},"agent_actions":{"view_html":"https://pith.science/pith/DDDQXKYHG27GELLBAY472UG6GB","download_json":"https://pith.science/pith/DDDQXKYHG27GELLBAY472UG6GB.json","view_paper":"https://pith.science/paper/DDDQXKYH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.13667&json=true","fetch_graph":"https://pith.science/api/pith-number/DDDQXKYHG27GELLBAY472UG6GB/graph.json","fetch_events":"https://pith.science/api/pith-number/DDDQXKYHG27GELLBAY472UG6GB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DDDQXKYHG27GELLBAY472UG6GB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DDDQXKYHG27GELLBAY472UG6GB/action/storage_attestation","attest_author":"https://pith.science/pith/DDDQXKYHG27GELLBAY472UG6GB/action/author_attestation","sign_citation":"https://pith.science/pith/DDDQXKYHG27GELLBAY472UG6GB/action/citation_signature","submit_replication":"https://pith.science/pith/DDDQXKYHG27GELLBAY472UG6GB/action/replication_record"}},"created_at":"2026-07-05T11:50:30.269305+00:00","updated_at":"2026-07-05T11:50:30.269305+00:00"}