{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:TZJ2KCQWCTXWBXHVHYB7223MGS","short_pith_number":"pith:TZJ2KCQW","schema_version":"1.0","canonical_sha256":"9e53a50a1614ef60dcf53e03fd6b6c349b78e92ba4d3d786072be5e99c3728de","source":{"kind":"arxiv","id":"2603.18943","version":2},"attestation_state":"computed","paper":{"title":"VGGT-360: Geometry-Consistent Zero-Shot Panoramic Depth Estimation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"VGGT-360 turns panoramic depth estimation into consistent 3D reprojection by leveraging VGGT foundation models without any training.","cross_cats":[],"primary_cat":"cs.CV","authors_text":"De Wen Soh, Haobo Jiang, Jiayi Yuan, Na Zhao","submitted_at":"2026-03-19T14:18:17Z","abstract_excerpt":"This paper presents VGGT-360, a novel training-free framework for zero-shot, geometry-consistent panoramic depth estimation. Unlike prior view-independent training-free approaches, VGGT-360 reformulates the task as panoramic reprojection over multi-view reconstructed 3D models by leveraging the intrinsic 3D consistency of VGGT-like foundation models, thereby unifying fragmented per-view reasoning into a coherent panoramic understanding. To achieve robust and accurate estimation, VGGT-360 integrates three plug-and-play modules that form a unified panorama-to-3D-to-depth framework: (i) Uncertain"},"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":true,"formal_links_present":false},"canonical_record":{"source":{"id":"2603.18943","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-03-19T14:18:17Z","cross_cats_sorted":[],"title_canon_sha256":"049151b81b25f7c0716f4c6edbd96d173665c76c46d6e4cbe73334223d841778","abstract_canon_sha256":"2afbc35d5fae08204959d7267c30842dea132b8402fdbaf9c19ec1627b922f8e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:39:15.733769Z","signature_b64":"p+BIEWHs5Ins8FTBWAZn+nSIpkhuPslN/m7uP0FxkcPGBOFSlyx8ya5uHwcDyB+AHQJtLEg7ApJSGR1MEjFjCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9e53a50a1614ef60dcf53e03fd6b6c349b78e92ba4d3d786072be5e99c3728de","last_reissued_at":"2026-05-17T23:39:15.733252Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:39:15.733252Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"VGGT-360: Geometry-Consistent Zero-Shot Panoramic Depth Estimation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"VGGT-360 turns panoramic depth estimation into consistent 3D reprojection by leveraging VGGT foundation models without any training.","cross_cats":[],"primary_cat":"cs.CV","authors_text":"De Wen Soh, Haobo Jiang, Jiayi Yuan, Na Zhao","submitted_at":"2026-03-19T14:18:17Z","abstract_excerpt":"This paper presents VGGT-360, a novel training-free framework for zero-shot, geometry-consistent panoramic depth estimation. Unlike prior view-independent training-free approaches, VGGT-360 reformulates the task as panoramic reprojection over multi-view reconstructed 3D models by leveraging the intrinsic 3D consistency of VGGT-like foundation models, thereby unifying fragmented per-view reasoning into a coherent panoramic understanding. To achieve robust and accurate estimation, VGGT-360 integrates three plug-and-play modules that form a unified panorama-to-3D-to-depth framework: (i) Uncertain"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"VGGT-360 outperforms both trained and training-free state-of-the-art methods across multiple resolutions and diverse indoor and outdoor datasets.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That VGGT-like foundation models possess intrinsic 3D consistency sufficient to unify fragmented per-view reasoning into coherent panoramic depth via the proposed reprojection pipeline.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"VGGT-360 delivers geometry-consistent zero-shot panoramic depth by converting panoramas into multi-view 3D reconstructions via VGGT models and three plug-and-play correction modules, then reprojecting the result.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"VGGT-360 turns panoramic depth estimation into consistent 3D reprojection by leveraging VGGT foundation models without any training.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"071bd4ba4b55ad5e3156d05724ff3bd6a7ad89e69ea8bc1227399db537f37da1"},"source":{"id":"2603.18943","kind":"arxiv","version":2},"verdict":{"id":"a931e22f-a09d-49b9-8e93-37be61f41b3f","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-15T08:34:42.567413Z","strongest_claim":"VGGT-360 outperforms both trained and training-free state-of-the-art methods across multiple resolutions and diverse indoor and outdoor datasets.","one_line_summary":"VGGT-360 delivers geometry-consistent zero-shot panoramic depth by converting panoramas into multi-view 3D reconstructions via VGGT models and three plug-and-play correction modules, then reprojecting the result.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That VGGT-like foundation models possess intrinsic 3D consistency sufficient to unify fragmented per-view reasoning into coherent panoramic depth via the proposed reprojection pipeline.","pith_extraction_headline":"VGGT-360 turns panoramic depth estimation into consistent 3D reprojection by leveraging VGGT foundation models without any training."},"references":{"count":52,"sample":[{"doi":"","year":2024,"title":"Elite360d: Towards efficient 360 depth estimation via semantic-and distance-aware bi- projection fusion","work_id":"5ae3d252-fc3e-418e-b4ea-edb860c6a30b","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"Hrdfuse: Monocular 360deg depth estimation by collaboratively learning holistic-with-regional depth distri- butions","work_id":"233a1acb-a5f3-4815-8eaa-fa2c1b5d6733","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2017,"title":"Joint 2d-3d-semantic data for indoor scene understanding","work_id":"cd49417b-13e3-4652-87f2-c992e78d093a","ref_index":3,"cited_arxiv_id":"1702.01105","is_internal_anchor":true},{"doi":"","year":2020,"title":"Matryodshka: Real-time 6dof video view synthesis using multi-sphere images","work_id":"e0436f48-5d89-4364-8f81-e7fad19b3280","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2020,"title":"Omniphotos: casual 360 vr photography.ACM Transactions on Graphics (TOG), 39(6):1–12, 2020","work_id":"06e8bd3c-d40a-404a-920f-023a3988c722","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":52,"snapshot_sha256":"85e94d8f8add17043ddeb9a80ce3aa0f2d539963560cd7f3992cc0bdbc2b2399","internal_anchors":6},"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":"2603.18943","created_at":"2026-05-17T23:39:15.733334+00:00"},{"alias_kind":"arxiv_version","alias_value":"2603.18943v2","created_at":"2026-05-17T23:39:15.733334+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.18943","created_at":"2026-05-17T23:39:15.733334+00:00"},{"alias_kind":"pith_short_12","alias_value":"TZJ2KCQWCTXW","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_16","alias_value":"TZJ2KCQWCTXWBXHV","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_8","alias_value":"TZJ2KCQW","created_at":"2026-05-18T12:33:37.589309+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2605.23892","citing_title":"Good Token Hunting: A Hitchhiker's Guide to Token Selection for Visual Geometry Transformers","ref_index":104,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TZJ2KCQWCTXWBXHVHYB7223MGS","json":"https://pith.science/pith/TZJ2KCQWCTXWBXHVHYB7223MGS.json","graph_json":"https://pith.science/api/pith-number/TZJ2KCQWCTXWBXHVHYB7223MGS/graph.json","events_json":"https://pith.science/api/pith-number/TZJ2KCQWCTXWBXHVHYB7223MGS/events.json","paper":"https://pith.science/paper/TZJ2KCQW"},"agent_actions":{"view_html":"https://pith.science/pith/TZJ2KCQWCTXWBXHVHYB7223MGS","download_json":"https://pith.science/pith/TZJ2KCQWCTXWBXHVHYB7223MGS.json","view_paper":"https://pith.science/paper/TZJ2KCQW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2603.18943&json=true","fetch_graph":"https://pith.science/api/pith-number/TZJ2KCQWCTXWBXHVHYB7223MGS/graph.json","fetch_events":"https://pith.science/api/pith-number/TZJ2KCQWCTXWBXHVHYB7223MGS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TZJ2KCQWCTXWBXHVHYB7223MGS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TZJ2KCQWCTXWBXHVHYB7223MGS/action/storage_attestation","attest_author":"https://pith.science/pith/TZJ2KCQWCTXWBXHVHYB7223MGS/action/author_attestation","sign_citation":"https://pith.science/pith/TZJ2KCQWCTXWBXHVHYB7223MGS/action/citation_signature","submit_replication":"https://pith.science/pith/TZJ2KCQWCTXWBXHVHYB7223MGS/action/replication_record"}},"created_at":"2026-05-17T23:39:15.733334+00:00","updated_at":"2026-05-17T23:39:15.733334+00:00"}