{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:YLBTCGHG5DMSKORFKWUENXMLC6","short_pith_number":"pith:YLBTCGHG","schema_version":"1.0","canonical_sha256":"c2c33118e6e8d9253a2555a846dd8b17bfc1423e7a68ba301462ce1f24a7fdd2","source":{"kind":"arxiv","id":"1904.08889","version":2},"attestation_state":"computed","paper":{"title":"KPConv: Flexible and Deformable Convolution for Point Clouds","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Beatriz Marcotegui, Charles R. Qi, Fran\\c{c}ois Goulette, Hugues Thomas, Jean-Emmanuel Deschaud, Leonidas J. Guibas","submitted_at":"2019-04-18T16:55:44Z","abstract_excerpt":"We present Kernel Point Convolution (KPConv), a new design of point convolution, i.e. that operates on point clouds without any intermediate representation. The convolution weights of KPConv are located in Euclidean space by kernel points, and applied to the input points close to them. Its capacity to use any number of kernel points gives KPConv more flexibility than fixed grid convolutions. Furthermore, these locations are continuous in space and can be learned by the network. Therefore, KPConv can be extended to deformable convolutions that learn to adapt kernel points to local geometry. Tha"},"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":"1904.08889","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2019-04-18T16:55:44Z","cross_cats_sorted":[],"title_canon_sha256":"2a0e93520f228a68071db839733c7fe50d9c5717865b2d290289e696ff433bc6","abstract_canon_sha256":"b94e7d9c0857f08e3fcfc3cc6a3364fbbb570b1a28436e71b370d9a2f9b110b6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-04T23:58:07.993762Z","signature_b64":"U2RmhdgJmeuYa2pPmn79k6l5vZvR/uOjO0zAq2c2S2rShFVAvKIUK59hp1cdOmnPNE5wfGrx6RONHMqVznQ5DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c2c33118e6e8d9253a2555a846dd8b17bfc1423e7a68ba301462ce1f24a7fdd2","last_reissued_at":"2026-07-04T23:58:07.993307Z","signature_status":"signed_v1","first_computed_at":"2026-07-04T23:58:07.993307Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"KPConv: Flexible and Deformable Convolution for Point Clouds","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Beatriz Marcotegui, Charles R. Qi, Fran\\c{c}ois Goulette, Hugues Thomas, Jean-Emmanuel Deschaud, Leonidas J. Guibas","submitted_at":"2019-04-18T16:55:44Z","abstract_excerpt":"We present Kernel Point Convolution (KPConv), a new design of point convolution, i.e. that operates on point clouds without any intermediate representation. The convolution weights of KPConv are located in Euclidean space by kernel points, and applied to the input points close to them. Its capacity to use any number of kernel points gives KPConv more flexibility than fixed grid convolutions. Furthermore, these locations are continuous in space and can be learned by the network. Therefore, KPConv can be extended to deformable convolutions that learn to adapt kernel points to local geometry. Tha"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1904.08889","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/1904.08889/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":"1904.08889","created_at":"2026-07-04T23:58:07.993376+00:00"},{"alias_kind":"arxiv_version","alias_value":"1904.08889v2","created_at":"2026-07-04T23:58:07.993376+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1904.08889","created_at":"2026-07-04T23:58:07.993376+00:00"},{"alias_kind":"pith_short_12","alias_value":"YLBTCGHG5DMS","created_at":"2026-07-04T23:58:07.993376+00:00"},{"alias_kind":"pith_short_16","alias_value":"YLBTCGHG5DMSKORF","created_at":"2026-07-04T23:58:07.993376+00:00"},{"alias_kind":"pith_short_8","alias_value":"YLBTCGHG","created_at":"2026-07-04T23:58:07.993376+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"1907.04444","citing_title":"A review on deep learning techniques for 3D sensed data classification","ref_index":97,"is_internal_anchor":false},{"citing_arxiv_id":"2605.02098","citing_title":"From Spherical to Gaussian: A Comparative Analysis of Point Cloud Cropping Strategies in Large-Scale 3D Environments","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2605.02098","citing_title":"From Spherical to Gaussian: A Comparative Analysis of Point Cloud Cropping Strategies in Large-Scale 3D Environments","ref_index":34,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YLBTCGHG5DMSKORFKWUENXMLC6","json":"https://pith.science/pith/YLBTCGHG5DMSKORFKWUENXMLC6.json","graph_json":"https://pith.science/api/pith-number/YLBTCGHG5DMSKORFKWUENXMLC6/graph.json","events_json":"https://pith.science/api/pith-number/YLBTCGHG5DMSKORFKWUENXMLC6/events.json","paper":"https://pith.science/paper/YLBTCGHG"},"agent_actions":{"view_html":"https://pith.science/pith/YLBTCGHG5DMSKORFKWUENXMLC6","download_json":"https://pith.science/pith/YLBTCGHG5DMSKORFKWUENXMLC6.json","view_paper":"https://pith.science/paper/YLBTCGHG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1904.08889&json=true","fetch_graph":"https://pith.science/api/pith-number/YLBTCGHG5DMSKORFKWUENXMLC6/graph.json","fetch_events":"https://pith.science/api/pith-number/YLBTCGHG5DMSKORFKWUENXMLC6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YLBTCGHG5DMSKORFKWUENXMLC6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YLBTCGHG5DMSKORFKWUENXMLC6/action/storage_attestation","attest_author":"https://pith.science/pith/YLBTCGHG5DMSKORFKWUENXMLC6/action/author_attestation","sign_citation":"https://pith.science/pith/YLBTCGHG5DMSKORFKWUENXMLC6/action/citation_signature","submit_replication":"https://pith.science/pith/YLBTCGHG5DMSKORFKWUENXMLC6/action/replication_record"}},"created_at":"2026-07-04T23:58:07.993376+00:00","updated_at":"2026-07-04T23:58:07.993376+00:00"}