{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:6XDW4JQBAXJQLPDMJOMGIPYV3G","short_pith_number":"pith:6XDW4JQB","schema_version":"1.0","canonical_sha256":"f5c76e260105d305bc6c4b98643f15d983089b8b0ad3f57350ca4e53f6f99f7c","source":{"kind":"arxiv","id":"2508.14036","version":2},"attestation_state":"computed","paper":{"title":"GeoSAM2: Unleashing the Power of SAM2 for 3D Part Segmentation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Ding Liang, Jingxiang Sun, Ken Deng, Xihui Liu, Yan-Pei Cao, Yebin Liu, Yunhan Yang","submitted_at":"2025-08-19T17:58:51Z","abstract_excerpt":"We introduce GeoSAM2, a prompt-controllable framework for 3D part segmentation that casts the task as multi-view 2D mask prediction. Given a textureless object, we render normal and point maps from predefined viewpoints and accept simple 2D prompts - clicks or boxes - to guide part selection. These prompts are processed by a shared SAM2 backbone augmented with LoRA and residual geometry fusion, enabling view-specific reasoning while preserving pretrained priors. The predicted masks are back-projected to the object and aggregated across views. Our method enables fine-grained, part-specific cont"},"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":"2508.14036","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-08-19T17:58:51Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"fe5baaa85fd6b4d12415de065f5ed7e544429bd5f1ad7dd0be7914670c169fa0","abstract_canon_sha256":"c8b54faae9689349b1249175913320d4f6c0fb750a2d7e9fbea4b3c2a3d8d1d3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:00:08.934840Z","signature_b64":"X3wXQlXyREHtI9MXFa44UywYjXz3e5YK7c2rs4GkI/UbaEyomlZ49AVogw8uax5+FB7681BkLeeBasFMdnXJBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f5c76e260105d305bc6c4b98643f15d983089b8b0ad3f57350ca4e53f6f99f7c","last_reissued_at":"2026-07-05T12:00:08.934392Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:00:08.934392Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"GeoSAM2: Unleashing the Power of SAM2 for 3D Part Segmentation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Ding Liang, Jingxiang Sun, Ken Deng, Xihui Liu, Yan-Pei Cao, Yebin Liu, Yunhan Yang","submitted_at":"2025-08-19T17:58:51Z","abstract_excerpt":"We introduce GeoSAM2, a prompt-controllable framework for 3D part segmentation that casts the task as multi-view 2D mask prediction. Given a textureless object, we render normal and point maps from predefined viewpoints and accept simple 2D prompts - clicks or boxes - to guide part selection. These prompts are processed by a shared SAM2 backbone augmented with LoRA and residual geometry fusion, enabling view-specific reasoning while preserving pretrained priors. The predicted masks are back-projected to the object and aggregated across views. Our method enables fine-grained, part-specific cont"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.14036","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/2508.14036/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":"2508.14036","created_at":"2026-07-05T12:00:08.934452+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.14036v2","created_at":"2026-07-05T12:00:08.934452+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.14036","created_at":"2026-07-05T12:00:08.934452+00:00"},{"alias_kind":"pith_short_12","alias_value":"6XDW4JQBAXJQ","created_at":"2026-07-05T12:00:08.934452+00:00"},{"alias_kind":"pith_short_16","alias_value":"6XDW4JQBAXJQLPDM","created_at":"2026-07-05T12:00:08.934452+00:00"},{"alias_kind":"pith_short_8","alias_value":"6XDW4JQB","created_at":"2026-07-05T12:00:08.934452+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.05070","citing_title":"Part-Level 3D Gaussian Vehicle Generation with Joint and Hinge Axis Estimation","ref_index":13,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6XDW4JQBAXJQLPDMJOMGIPYV3G","json":"https://pith.science/pith/6XDW4JQBAXJQLPDMJOMGIPYV3G.json","graph_json":"https://pith.science/api/pith-number/6XDW4JQBAXJQLPDMJOMGIPYV3G/graph.json","events_json":"https://pith.science/api/pith-number/6XDW4JQBAXJQLPDMJOMGIPYV3G/events.json","paper":"https://pith.science/paper/6XDW4JQB"},"agent_actions":{"view_html":"https://pith.science/pith/6XDW4JQBAXJQLPDMJOMGIPYV3G","download_json":"https://pith.science/pith/6XDW4JQBAXJQLPDMJOMGIPYV3G.json","view_paper":"https://pith.science/paper/6XDW4JQB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.14036&json=true","fetch_graph":"https://pith.science/api/pith-number/6XDW4JQBAXJQLPDMJOMGIPYV3G/graph.json","fetch_events":"https://pith.science/api/pith-number/6XDW4JQBAXJQLPDMJOMGIPYV3G/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6XDW4JQBAXJQLPDMJOMGIPYV3G/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6XDW4JQBAXJQLPDMJOMGIPYV3G/action/storage_attestation","attest_author":"https://pith.science/pith/6XDW4JQBAXJQLPDMJOMGIPYV3G/action/author_attestation","sign_citation":"https://pith.science/pith/6XDW4JQBAXJQLPDMJOMGIPYV3G/action/citation_signature","submit_replication":"https://pith.science/pith/6XDW4JQBAXJQLPDMJOMGIPYV3G/action/replication_record"}},"created_at":"2026-07-05T12:00:08.934452+00:00","updated_at":"2026-07-05T12:00:08.934452+00:00"}