{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:I4OLFVIJPD5CGLGROTZYGX2KGO","short_pith_number":"pith:I4OLFVIJ","schema_version":"1.0","canonical_sha256":"471cb2d50978fa232cd174f3835f4a3387c481ea8f04f843a36e201d8286019b","source":{"kind":"arxiv","id":"2309.14491","version":1},"attestation_state":"computed","paper":{"title":"Unsupervised 3D Perception with 2D Vision-Language Distillation for Autonomous Driving","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Charles R. Qi, Dragomir Anguelov, Jingwei Ji, Mahyar Najibi, Scott Ettinger, Xinchen Yan, Yin Zhou","submitted_at":"2023-09-25T19:33:52Z","abstract_excerpt":"Closed-set 3D perception models trained on only a pre-defined set of object categories can be inadequate for safety critical applications such as autonomous driving where new object types can be encountered after deployment. In this paper, we present a multi-modal auto labeling pipeline capable of generating amodal 3D bounding boxes and tracklets for training models on open-set categories without 3D human labels. Our pipeline exploits motion cues inherent in point cloud sequences in combination with the freely available 2D image-text pairs to identify and track all traffic participants. Compar"},"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":"2309.14491","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-09-25T19:33:52Z","cross_cats_sorted":[],"title_canon_sha256":"cdf8bba5580eb4f4e356d1d1e265e4154becbf69ba6ac2777814e96ecf892fe6","abstract_canon_sha256":"c69a2547ca099adf0b78ae1ee07902130c2096bad1dfabf3b328cf928588abb6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:54:26.198864Z","signature_b64":"IobG2IcG13b2D67g1XADebhzkvhqVcJNqwKGd6ES1f8e42j/hqcjR8oCUmLRg29F8grAqaZ7vwIAGfzAe2qOAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"471cb2d50978fa232cd174f3835f4a3387c481ea8f04f843a36e201d8286019b","last_reissued_at":"2026-07-05T06:54:26.198400Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:54:26.198400Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Unsupervised 3D Perception with 2D Vision-Language Distillation for Autonomous Driving","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Charles R. Qi, Dragomir Anguelov, Jingwei Ji, Mahyar Najibi, Scott Ettinger, Xinchen Yan, Yin Zhou","submitted_at":"2023-09-25T19:33:52Z","abstract_excerpt":"Closed-set 3D perception models trained on only a pre-defined set of object categories can be inadequate for safety critical applications such as autonomous driving where new object types can be encountered after deployment. In this paper, we present a multi-modal auto labeling pipeline capable of generating amodal 3D bounding boxes and tracklets for training models on open-set categories without 3D human labels. Our pipeline exploits motion cues inherent in point cloud sequences in combination with the freely available 2D image-text pairs to identify and track all traffic participants. Compar"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2309.14491","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/2309.14491/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":"2309.14491","created_at":"2026-07-05T06:54:26.198457+00:00"},{"alias_kind":"arxiv_version","alias_value":"2309.14491v1","created_at":"2026-07-05T06:54:26.198457+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2309.14491","created_at":"2026-07-05T06:54:26.198457+00:00"},{"alias_kind":"pith_short_12","alias_value":"I4OLFVIJPD5C","created_at":"2026-07-05T06:54:26.198457+00:00"},{"alias_kind":"pith_short_16","alias_value":"I4OLFVIJPD5CGLGR","created_at":"2026-07-05T06:54:26.198457+00:00"},{"alias_kind":"pith_short_8","alias_value":"I4OLFVIJ","created_at":"2026-07-05T06:54:26.198457+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2502.03901","citing_title":"LeAP: Consistent multi-domain 3D labeling using Foundation Models","ref_index":62,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/I4OLFVIJPD5CGLGROTZYGX2KGO","json":"https://pith.science/pith/I4OLFVIJPD5CGLGROTZYGX2KGO.json","graph_json":"https://pith.science/api/pith-number/I4OLFVIJPD5CGLGROTZYGX2KGO/graph.json","events_json":"https://pith.science/api/pith-number/I4OLFVIJPD5CGLGROTZYGX2KGO/events.json","paper":"https://pith.science/paper/I4OLFVIJ"},"agent_actions":{"view_html":"https://pith.science/pith/I4OLFVIJPD5CGLGROTZYGX2KGO","download_json":"https://pith.science/pith/I4OLFVIJPD5CGLGROTZYGX2KGO.json","view_paper":"https://pith.science/paper/I4OLFVIJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2309.14491&json=true","fetch_graph":"https://pith.science/api/pith-number/I4OLFVIJPD5CGLGROTZYGX2KGO/graph.json","fetch_events":"https://pith.science/api/pith-number/I4OLFVIJPD5CGLGROTZYGX2KGO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/I4OLFVIJPD5CGLGROTZYGX2KGO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/I4OLFVIJPD5CGLGROTZYGX2KGO/action/storage_attestation","attest_author":"https://pith.science/pith/I4OLFVIJPD5CGLGROTZYGX2KGO/action/author_attestation","sign_citation":"https://pith.science/pith/I4OLFVIJPD5CGLGROTZYGX2KGO/action/citation_signature","submit_replication":"https://pith.science/pith/I4OLFVIJPD5CGLGROTZYGX2KGO/action/replication_record"}},"created_at":"2026-07-05T06:54:26.198457+00:00","updated_at":"2026-07-05T06:54:26.198457+00:00"}