{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:3OEWWSMYJCNY7P6NU2QIGCWYVP","short_pith_number":"pith:3OEWWSMY","schema_version":"1.0","canonical_sha256":"db896b4998489b8fbfcda6a0830ad8abc20f2a8b66fbeb204bf7e587a67dda01","source":{"kind":"arxiv","id":"2606.26087","version":1},"attestation_state":"computed","paper":{"title":"MVTrack4Gen: Multi-View Point Tracking as Geometric Supervision for 4D Video Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Hyunsung Kim, Jaewoo Jung, Jahyeok Koo, Jisang Han, Jongmin Lee, JoungBin Lee, Kazumi Fukuda, Seungryong Kim, Takuya Narihira, Tongmin Kim, Yuki Mitsufuji","submitted_at":"2026-06-24T17:56:33Z","abstract_excerpt":"Synthesizing a novel-view video from a monocular reference video along a target camera trajectory requires both geometric consistency and motion fidelity with respect to the reference video. Existing methods based on explicit 3D representations are limited by the accuracy of off-the-shelf reconstruction modules, which often produce inaccurate geometry for dynamic objects in monocular videos. In contrast, camera-conditioning-only methods can achieve high visual quality but often struggle to preserve geometric and motion consistency. In this work, we introduce MVTrack4Gen (Multi-View point Track"},"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":"2606.26087","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-06-24T17:56:33Z","cross_cats_sorted":[],"title_canon_sha256":"eee337522030e7610d0fe476b8734eb34c78683ceaf91ba259e58fe2b39d3025","abstract_canon_sha256":"0ce62fee4e2215bd5e7c6c3e85f4588e549bc0967ef0b784d3debb464221e678"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-25T01:18:48.201085Z","signature_b64":"nBCYHzF5RlCI/WExTdG3DfQn3kkSFZ7xoo2kRxQd/jVFWTwmGqnL4hAKSIRJWZKuBDGDaCYaHPHyn/A+7rTmDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"db896b4998489b8fbfcda6a0830ad8abc20f2a8b66fbeb204bf7e587a67dda01","last_reissued_at":"2026-06-25T01:18:48.200596Z","signature_status":"signed_v1","first_computed_at":"2026-06-25T01:18:48.200596Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MVTrack4Gen: Multi-View Point Tracking as Geometric Supervision for 4D Video Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Hyunsung Kim, Jaewoo Jung, Jahyeok Koo, Jisang Han, Jongmin Lee, JoungBin Lee, Kazumi Fukuda, Seungryong Kim, Takuya Narihira, Tongmin Kim, Yuki Mitsufuji","submitted_at":"2026-06-24T17:56:33Z","abstract_excerpt":"Synthesizing a novel-view video from a monocular reference video along a target camera trajectory requires both geometric consistency and motion fidelity with respect to the reference video. Existing methods based on explicit 3D representations are limited by the accuracy of off-the-shelf reconstruction modules, which often produce inaccurate geometry for dynamic objects in monocular videos. In contrast, camera-conditioning-only methods can achieve high visual quality but often struggle to preserve geometric and motion consistency. In this work, we introduce MVTrack4Gen (Multi-View point Track"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.26087","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/2606.26087/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":"2606.26087","created_at":"2026-06-25T01:18:48.200661+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.26087v1","created_at":"2026-06-25T01:18:48.200661+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.26087","created_at":"2026-06-25T01:18:48.200661+00:00"},{"alias_kind":"pith_short_12","alias_value":"3OEWWSMYJCNY","created_at":"2026-06-25T01:18:48.200661+00:00"},{"alias_kind":"pith_short_16","alias_value":"3OEWWSMYJCNY7P6N","created_at":"2026-06-25T01:18:48.200661+00:00"},{"alias_kind":"pith_short_8","alias_value":"3OEWWSMY","created_at":"2026-06-25T01:18:48.200661+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/3OEWWSMYJCNY7P6NU2QIGCWYVP","json":"https://pith.science/pith/3OEWWSMYJCNY7P6NU2QIGCWYVP.json","graph_json":"https://pith.science/api/pith-number/3OEWWSMYJCNY7P6NU2QIGCWYVP/graph.json","events_json":"https://pith.science/api/pith-number/3OEWWSMYJCNY7P6NU2QIGCWYVP/events.json","paper":"https://pith.science/paper/3OEWWSMY"},"agent_actions":{"view_html":"https://pith.science/pith/3OEWWSMYJCNY7P6NU2QIGCWYVP","download_json":"https://pith.science/pith/3OEWWSMYJCNY7P6NU2QIGCWYVP.json","view_paper":"https://pith.science/paper/3OEWWSMY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.26087&json=true","fetch_graph":"https://pith.science/api/pith-number/3OEWWSMYJCNY7P6NU2QIGCWYVP/graph.json","fetch_events":"https://pith.science/api/pith-number/3OEWWSMYJCNY7P6NU2QIGCWYVP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3OEWWSMYJCNY7P6NU2QIGCWYVP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3OEWWSMYJCNY7P6NU2QIGCWYVP/action/storage_attestation","attest_author":"https://pith.science/pith/3OEWWSMYJCNY7P6NU2QIGCWYVP/action/author_attestation","sign_citation":"https://pith.science/pith/3OEWWSMYJCNY7P6NU2QIGCWYVP/action/citation_signature","submit_replication":"https://pith.science/pith/3OEWWSMYJCNY7P6NU2QIGCWYVP/action/replication_record"}},"created_at":"2026-06-25T01:18:48.200661+00:00","updated_at":"2026-06-25T01:18:48.200661+00:00"}