{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:TPOE33AZCYW5OEWFB6JVNENWOY","short_pith_number":"pith:TPOE33AZ","schema_version":"1.0","canonical_sha256":"9bdc4dec19162dd712c50f935691b676069a8dd56c252b74770bc8a1bf4b585b","source":{"kind":"arxiv","id":"2606.00299","version":1},"attestation_state":"computed","paper":{"title":"Real2SAM2Real: Generative 3D Caches as Complementary Context for Video Diffusion","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Christopher Metzler, Cornelia Fermuller, Haoming Cai, Jiayi Wu, Yiannis Aloimonos","submitted_at":"2026-05-29T19:28:07Z","abstract_excerpt":"While Video Diffusion Models (VDMs) excel at synthesizing high-fidelity videos, enabling precise camera and scene control remains challenging. Existing methods predominantly rely on implicit diffusion priors to generate unobserved regions, inevitably leading to structural collapse during high-dynamic movements or complex occlusions. To address this challenge, we propose Real2SAM2Real, a framework that leverages 3D lifting models (e.g., SAM3D) to extract an explicitly editable 3D cache, serving as a robust geometric scaffold for the VDM. By capturing the entire 3D volume of foreground entities "},"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":"2606.00299","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-29T19:28:07Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"047a4eaeb1280d19bef2d53eefe84e3331f46d1f2127762ca76a22c162d7d722","abstract_canon_sha256":"7f62f718af006a791fb52f30cf9a7f4c2cfb2fcf3aab3eb0e3f7b0c0a59a5df5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-02T01:03:51.023838Z","signature_b64":"aA7RPiSPNhRQZ96JuDvcvQbpbmEJo5jiKc8i7SDO9H/t41Qxf1WSZd9hP/HKLJpW1JbLjU7b1yX0z6lB3gRTAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9bdc4dec19162dd712c50f935691b676069a8dd56c252b74770bc8a1bf4b585b","last_reissued_at":"2026-06-02T01:03:51.023361Z","signature_status":"signed_v1","first_computed_at":"2026-06-02T01:03:51.023361Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Real2SAM2Real: Generative 3D Caches as Complementary Context for Video Diffusion","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Christopher Metzler, Cornelia Fermuller, Haoming Cai, Jiayi Wu, Yiannis Aloimonos","submitted_at":"2026-05-29T19:28:07Z","abstract_excerpt":"While Video Diffusion Models (VDMs) excel at synthesizing high-fidelity videos, enabling precise camera and scene control remains challenging. Existing methods predominantly rely on implicit diffusion priors to generate unobserved regions, inevitably leading to structural collapse during high-dynamic movements or complex occlusions. To address this challenge, we propose Real2SAM2Real, a framework that leverages 3D lifting models (e.g., SAM3D) to extract an explicitly editable 3D cache, serving as a robust geometric scaffold for the VDM. By capturing the entire 3D volume of foreground entities "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.00299","kind":"arxiv","version":1},"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/2606.00299/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":"2606.00299","created_at":"2026-06-02T01:03:51.023421+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.00299v1","created_at":"2026-06-02T01:03:51.023421+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.00299","created_at":"2026-06-02T01:03:51.023421+00:00"},{"alias_kind":"pith_short_12","alias_value":"TPOE33AZCYW5","created_at":"2026-06-02T01:03:51.023421+00:00"},{"alias_kind":"pith_short_16","alias_value":"TPOE33AZCYW5OEWF","created_at":"2026-06-02T01:03:51.023421+00:00"},{"alias_kind":"pith_short_8","alias_value":"TPOE33AZ","created_at":"2026-06-02T01:03:51.023421+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/TPOE33AZCYW5OEWFB6JVNENWOY","json":"https://pith.science/pith/TPOE33AZCYW5OEWFB6JVNENWOY.json","graph_json":"https://pith.science/api/pith-number/TPOE33AZCYW5OEWFB6JVNENWOY/graph.json","events_json":"https://pith.science/api/pith-number/TPOE33AZCYW5OEWFB6JVNENWOY/events.json","paper":"https://pith.science/paper/TPOE33AZ"},"agent_actions":{"view_html":"https://pith.science/pith/TPOE33AZCYW5OEWFB6JVNENWOY","download_json":"https://pith.science/pith/TPOE33AZCYW5OEWFB6JVNENWOY.json","view_paper":"https://pith.science/paper/TPOE33AZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.00299&json=true","fetch_graph":"https://pith.science/api/pith-number/TPOE33AZCYW5OEWFB6JVNENWOY/graph.json","fetch_events":"https://pith.science/api/pith-number/TPOE33AZCYW5OEWFB6JVNENWOY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TPOE33AZCYW5OEWFB6JVNENWOY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TPOE33AZCYW5OEWFB6JVNENWOY/action/storage_attestation","attest_author":"https://pith.science/pith/TPOE33AZCYW5OEWFB6JVNENWOY/action/author_attestation","sign_citation":"https://pith.science/pith/TPOE33AZCYW5OEWFB6JVNENWOY/action/citation_signature","submit_replication":"https://pith.science/pith/TPOE33AZCYW5OEWFB6JVNENWOY/action/replication_record"}},"created_at":"2026-06-02T01:03:51.023421+00:00","updated_at":"2026-06-02T01:03:51.023421+00:00"}