{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:HZ3PAL3F2XGCYAIINVJBNCWEJ7","short_pith_number":"pith:HZ3PAL3F","schema_version":"1.0","canonical_sha256":"3e76f02f65d5cc2c01086d52168ac44fc232b4855c225814583616fdfd2f84dd","source":{"kind":"arxiv","id":"2306.07279","version":2},"attestation_state":"computed","paper":{"title":"Scalable 3D Captioning with Pretrained Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chris Rockwell, Honglak Lee, Justin Johnson, Tiange Luo","submitted_at":"2023-06-12T17:59:03Z","abstract_excerpt":"We introduce Cap3D, an automatic approach for generating descriptive text for 3D objects. This approach utilizes pretrained models from image captioning, image-text alignment, and LLM to consolidate captions from multiple views of a 3D asset, completely side-stepping the time-consuming and costly process of manual annotation. We apply Cap3D to the recently introduced large-scale 3D dataset, Objaverse, resulting in 660k 3D-text pairs. Our evaluation, conducted using 41k human annotations from the same dataset, demonstrates that Cap3D surpasses human-authored descriptions in terms of quality, co"},"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":"2306.07279","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-06-12T17:59:03Z","cross_cats_sorted":[],"title_canon_sha256":"f670d75e3bce2ba7414ff62e8b332c4c8e3d6dc97a674ae51cc617cce3d67f27","abstract_canon_sha256":"36995ec1ecd3e70f92ed5a692eb0a4b5659cac0edad71a96af5047042156b768"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:21:17.952790Z","signature_b64":"rmrx93dMfd9k5LktBW6YILM4vH0kteN2W4cSOlcyMOD0y6RLYlTkgSrczV/s4qQRb3DZxrzXHs9xMUf54eFuCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3e76f02f65d5cc2c01086d52168ac44fc232b4855c225814583616fdfd2f84dd","last_reissued_at":"2026-07-05T06:21:17.952293Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:21:17.952293Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Scalable 3D Captioning with Pretrained Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chris Rockwell, Honglak Lee, Justin Johnson, Tiange Luo","submitted_at":"2023-06-12T17:59:03Z","abstract_excerpt":"We introduce Cap3D, an automatic approach for generating descriptive text for 3D objects. This approach utilizes pretrained models from image captioning, image-text alignment, and LLM to consolidate captions from multiple views of a 3D asset, completely side-stepping the time-consuming and costly process of manual annotation. We apply Cap3D to the recently introduced large-scale 3D dataset, Objaverse, resulting in 660k 3D-text pairs. Our evaluation, conducted using 41k human annotations from the same dataset, demonstrates that Cap3D surpasses human-authored descriptions in terms of quality, co"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.07279","kind":"arxiv","version":2},"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/2306.07279/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":"2306.07279","created_at":"2026-07-05T06:21:17.952359+00:00"},{"alias_kind":"arxiv_version","alias_value":"2306.07279v2","created_at":"2026-07-05T06:21:17.952359+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.07279","created_at":"2026-07-05T06:21:17.952359+00:00"},{"alias_kind":"pith_short_12","alias_value":"HZ3PAL3F2XGC","created_at":"2026-07-05T06:21:17.952359+00:00"},{"alias_kind":"pith_short_16","alias_value":"HZ3PAL3F2XGCYAII","created_at":"2026-07-05T06:21:17.952359+00:00"},{"alias_kind":"pith_short_8","alias_value":"HZ3PAL3F","created_at":"2026-07-05T06:21:17.952359+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2607.01803","citing_title":"PixGS: Pixel-Space Diffusion for Direct 3D Gaussian Splat Generation","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2607.00522","citing_title":"Restore3D: Breathing Life into Broken Objects with Shape and Texture Restoration","ref_index":84,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18052","citing_title":"Efficient 3D Content Reconstruction and Generation","ref_index":150,"is_internal_anchor":false},{"citing_arxiv_id":"2509.10813","citing_title":"InternScenes: A Large-scale Simulatable Indoor Scene Dataset with Realistic Layouts","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2311.04400","citing_title":"LRM: Large Reconstruction Model for Single Image to 3D","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2407.07726","citing_title":"PaliGemma: A versatile 3B VLM for transfer","ref_index":89,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HZ3PAL3F2XGCYAIINVJBNCWEJ7","json":"https://pith.science/pith/HZ3PAL3F2XGCYAIINVJBNCWEJ7.json","graph_json":"https://pith.science/api/pith-number/HZ3PAL3F2XGCYAIINVJBNCWEJ7/graph.json","events_json":"https://pith.science/api/pith-number/HZ3PAL3F2XGCYAIINVJBNCWEJ7/events.json","paper":"https://pith.science/paper/HZ3PAL3F"},"agent_actions":{"view_html":"https://pith.science/pith/HZ3PAL3F2XGCYAIINVJBNCWEJ7","download_json":"https://pith.science/pith/HZ3PAL3F2XGCYAIINVJBNCWEJ7.json","view_paper":"https://pith.science/paper/HZ3PAL3F","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2306.07279&json=true","fetch_graph":"https://pith.science/api/pith-number/HZ3PAL3F2XGCYAIINVJBNCWEJ7/graph.json","fetch_events":"https://pith.science/api/pith-number/HZ3PAL3F2XGCYAIINVJBNCWEJ7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HZ3PAL3F2XGCYAIINVJBNCWEJ7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HZ3PAL3F2XGCYAIINVJBNCWEJ7/action/storage_attestation","attest_author":"https://pith.science/pith/HZ3PAL3F2XGCYAIINVJBNCWEJ7/action/author_attestation","sign_citation":"https://pith.science/pith/HZ3PAL3F2XGCYAIINVJBNCWEJ7/action/citation_signature","submit_replication":"https://pith.science/pith/HZ3PAL3F2XGCYAIINVJBNCWEJ7/action/replication_record"}},"created_at":"2026-07-05T06:21:17.952359+00:00","updated_at":"2026-07-05T06:21:17.952359+00:00"}