{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:RATDRQDEFAIOF2YY6XJLGN4F72","short_pith_number":"pith:RATDRQDE","schema_version":"1.0","canonical_sha256":"882638c0642810e2eb18f5d2b33785fead9dec95e9fa70de136639f6c116fe4a","source":{"kind":"arxiv","id":"2402.05235","version":1},"attestation_state":"computed","paper":{"title":"SPAD : Spatially Aware Multiview Diffusers","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Aliaksandr Siarohin, Bernard Ghanem, Guocheng Qian, Igor Gilitschenski, Jian Ren, Michael Vasilkovsky, Riza Alp Guler, Sergey Tulyakov, Yash Kant, Ziyi Wu","submitted_at":"2024-02-07T20:16:09Z","abstract_excerpt":"We present SPAD, a novel approach for creating consistent multi-view images from text prompts or single images. To enable multi-view generation, we repurpose a pretrained 2D diffusion model by extending its self-attention layers with cross-view interactions, and fine-tune it on a high quality subset of Objaverse. We find that a naive extension of the self-attention proposed in prior work (e.g. MVDream) leads to content copying between views. Therefore, we explicitly constrain the cross-view attention based on epipolar geometry. To further enhance 3D consistency, we utilize Plucker coordinates "},"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":"2402.05235","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-02-07T20:16:09Z","cross_cats_sorted":[],"title_canon_sha256":"e7300e0592deb43b82bc17b868e88bd789dccaf112650c3452797365b0648945","abstract_canon_sha256":"6b1dc3fe05ae98d3c7edecd29117e1efb8e29f395212934cae7d5db7ebb5aebf"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:42:46.543429Z","signature_b64":"q7sQWWBBX6k+2JCYb/BDGUvLCZst/BhQG9i9HmqNFno15fmSzcyBJeRzh+uNJB0dOFxK85Of9F3gHwoa6QAnDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"882638c0642810e2eb18f5d2b33785fead9dec95e9fa70de136639f6c116fe4a","last_reissued_at":"2026-07-05T07:42:46.543043Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:42:46.543043Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SPAD : Spatially Aware Multiview Diffusers","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Aliaksandr Siarohin, Bernard Ghanem, Guocheng Qian, Igor Gilitschenski, Jian Ren, Michael Vasilkovsky, Riza Alp Guler, Sergey Tulyakov, Yash Kant, Ziyi Wu","submitted_at":"2024-02-07T20:16:09Z","abstract_excerpt":"We present SPAD, a novel approach for creating consistent multi-view images from text prompts or single images. To enable multi-view generation, we repurpose a pretrained 2D diffusion model by extending its self-attention layers with cross-view interactions, and fine-tune it on a high quality subset of Objaverse. We find that a naive extension of the self-attention proposed in prior work (e.g. MVDream) leads to content copying between views. Therefore, we explicitly constrain the cross-view attention based on epipolar geometry. To further enhance 3D consistency, we utilize Plucker coordinates "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.05235","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/2402.05235/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":"2402.05235","created_at":"2026-07-05T07:42:46.543100+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.05235v1","created_at":"2026-07-05T07:42:46.543100+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.05235","created_at":"2026-07-05T07:42:46.543100+00:00"},{"alias_kind":"pith_short_12","alias_value":"RATDRQDEFAIO","created_at":"2026-07-05T07:42:46.543100+00:00"},{"alias_kind":"pith_short_16","alias_value":"RATDRQDEFAIOF2YY","created_at":"2026-07-05T07:42:46.543100+00:00"},{"alias_kind":"pith_short_8","alias_value":"RATDRQDE","created_at":"2026-07-05T07:42:46.543100+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2406.02509","citing_title":"CamCo: Camera-Controllable 3D-Consistent Image-to-Video Generation","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2404.07191","citing_title":"InstantMesh: Efficient 3D Mesh Generation from a Single Image with Sparse-view Large Reconstruction Models","ref_index":17,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RATDRQDEFAIOF2YY6XJLGN4F72","json":"https://pith.science/pith/RATDRQDEFAIOF2YY6XJLGN4F72.json","graph_json":"https://pith.science/api/pith-number/RATDRQDEFAIOF2YY6XJLGN4F72/graph.json","events_json":"https://pith.science/api/pith-number/RATDRQDEFAIOF2YY6XJLGN4F72/events.json","paper":"https://pith.science/paper/RATDRQDE"},"agent_actions":{"view_html":"https://pith.science/pith/RATDRQDEFAIOF2YY6XJLGN4F72","download_json":"https://pith.science/pith/RATDRQDEFAIOF2YY6XJLGN4F72.json","view_paper":"https://pith.science/paper/RATDRQDE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.05235&json=true","fetch_graph":"https://pith.science/api/pith-number/RATDRQDEFAIOF2YY6XJLGN4F72/graph.json","fetch_events":"https://pith.science/api/pith-number/RATDRQDEFAIOF2YY6XJLGN4F72/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RATDRQDEFAIOF2YY6XJLGN4F72/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RATDRQDEFAIOF2YY6XJLGN4F72/action/storage_attestation","attest_author":"https://pith.science/pith/RATDRQDEFAIOF2YY6XJLGN4F72/action/author_attestation","sign_citation":"https://pith.science/pith/RATDRQDEFAIOF2YY6XJLGN4F72/action/citation_signature","submit_replication":"https://pith.science/pith/RATDRQDEFAIOF2YY6XJLGN4F72/action/replication_record"}},"created_at":"2026-07-05T07:42:46.543100+00:00","updated_at":"2026-07-05T07:42:46.543100+00:00"}