{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:P63C7SMMURZMKY2SL4O5VNHW5C","short_pith_number":"pith:P63C7SMM","schema_version":"1.0","canonical_sha256":"7fb62fc98ca472c563525f1ddab4f6e8b0d8dfbf5a7cddb3c4af5bf71c015242","source":{"kind":"arxiv","id":"2505.05473","version":1},"attestation_state":"computed","paper":{"title":"DiffusionSfM: Predicting Structure and Motion via Ray Origin and Endpoint Diffusion","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Amy Lin, Deva Ramanan, Jason Y. Zhang, Jeff Tan, Qitao Zhao, Shubham Tulsiani","submitted_at":"2025-05-08T17:59:47Z","abstract_excerpt":"Current Structure-from-Motion (SfM) methods typically follow a two-stage pipeline, combining learned or geometric pairwise reasoning with a subsequent global optimization step. In contrast, we propose a data-driven multi-view reasoning approach that directly infers 3D scene geometry and camera poses from multi-view images. Our framework, DiffusionSfM, parameterizes scene geometry and cameras as pixel-wise ray origins and endpoints in a global frame and employs a transformer-based denoising diffusion model to predict them from multi-view inputs. To address practical challenges in training diffu"},"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":"2505.05473","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2025-05-08T17:59:47Z","cross_cats_sorted":[],"title_canon_sha256":"d96196dff41ecdf9e8105ae8337b1c5b4669432e3e9ea50a12949e198adfedf3","abstract_canon_sha256":"65b089fb0170753034d8dd5f6e1c06f844b77ba500d8fd18ac0bd936dc9d5b24"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:00:24.768527Z","signature_b64":"urtkiYNNtxw8Fz8ebs4JvQvL5w9qRDBaFBUSdAtBej8+WeNZ3NgK9xgnXFT5/JzasP6443KCysk/SOT6AzFyCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7fb62fc98ca472c563525f1ddab4f6e8b0d8dfbf5a7cddb3c4af5bf71c015242","last_reissued_at":"2026-07-05T11:00:24.767988Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:00:24.767988Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DiffusionSfM: Predicting Structure and Motion via Ray Origin and Endpoint Diffusion","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Amy Lin, Deva Ramanan, Jason Y. Zhang, Jeff Tan, Qitao Zhao, Shubham Tulsiani","submitted_at":"2025-05-08T17:59:47Z","abstract_excerpt":"Current Structure-from-Motion (SfM) methods typically follow a two-stage pipeline, combining learned or geometric pairwise reasoning with a subsequent global optimization step. In contrast, we propose a data-driven multi-view reasoning approach that directly infers 3D scene geometry and camera poses from multi-view images. Our framework, DiffusionSfM, parameterizes scene geometry and cameras as pixel-wise ray origins and endpoints in a global frame and employs a transformer-based denoising diffusion model to predict them from multi-view inputs. To address practical challenges in training diffu"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.05473","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/2505.05473/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":"2505.05473","created_at":"2026-07-05T11:00:24.768056+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.05473v1","created_at":"2026-07-05T11:00:24.768056+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.05473","created_at":"2026-07-05T11:00:24.768056+00:00"},{"alias_kind":"pith_short_12","alias_value":"P63C7SMMURZM","created_at":"2026-07-05T11:00:24.768056+00:00"},{"alias_kind":"pith_short_16","alias_value":"P63C7SMMURZMKY2S","created_at":"2026-07-05T11:00:24.768056+00:00"},{"alias_kind":"pith_short_8","alias_value":"P63C7SMM","created_at":"2026-07-05T11:00:24.768056+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2508.10934","citing_title":"ViPE: Video Pose Engine for 3D Geometric Perception","ref_index":89,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/P63C7SMMURZMKY2SL4O5VNHW5C","json":"https://pith.science/pith/P63C7SMMURZMKY2SL4O5VNHW5C.json","graph_json":"https://pith.science/api/pith-number/P63C7SMMURZMKY2SL4O5VNHW5C/graph.json","events_json":"https://pith.science/api/pith-number/P63C7SMMURZMKY2SL4O5VNHW5C/events.json","paper":"https://pith.science/paper/P63C7SMM"},"agent_actions":{"view_html":"https://pith.science/pith/P63C7SMMURZMKY2SL4O5VNHW5C","download_json":"https://pith.science/pith/P63C7SMMURZMKY2SL4O5VNHW5C.json","view_paper":"https://pith.science/paper/P63C7SMM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.05473&json=true","fetch_graph":"https://pith.science/api/pith-number/P63C7SMMURZMKY2SL4O5VNHW5C/graph.json","fetch_events":"https://pith.science/api/pith-number/P63C7SMMURZMKY2SL4O5VNHW5C/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/P63C7SMMURZMKY2SL4O5VNHW5C/action/timestamp_anchor","attest_storage":"https://pith.science/pith/P63C7SMMURZMKY2SL4O5VNHW5C/action/storage_attestation","attest_author":"https://pith.science/pith/P63C7SMMURZMKY2SL4O5VNHW5C/action/author_attestation","sign_citation":"https://pith.science/pith/P63C7SMMURZMKY2SL4O5VNHW5C/action/citation_signature","submit_replication":"https://pith.science/pith/P63C7SMMURZMKY2SL4O5VNHW5C/action/replication_record"}},"created_at":"2026-07-05T11:00:24.768056+00:00","updated_at":"2026-07-05T11:00:24.768056+00:00"}