{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:PDUMO5PEW3MR6I6LWOVVVATYOL","short_pith_number":"pith:PDUMO5PE","schema_version":"1.0","canonical_sha256":"78e8c775e4b6d91f23cbb3ab5a827872f2a77c00bb2357e9c5c7461c7ac0bab5","source":{"kind":"arxiv","id":"2503.18947","version":1},"attestation_state":"computed","paper":{"title":"Tuning-Free Amodal Segmentation via the Occlusion-Free Bias of Inpainting Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bedrich Benes, Jae Joong Lee, Raymond A. Yeh","submitted_at":"2025-03-24T17:59:56Z","abstract_excerpt":"Amodal segmentation aims to predict segmentation masks for both the visible and occluded regions of an object. Most existing works formulate this as a supervised learning problem, requiring manually annotated amodal masks or synthetic training data. Consequently, their performance depends on the quality of the datasets, which often lack diversity and scale. This work introduces a tuning-free approach that repurposes pretrained diffusion-based inpainting models for amodal segmentation. Our approach is motivated by the \"occlusion-free bias\" of inpainting models, i.e., the inpainted objects tend "},"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":"2503.18947","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-03-24T17:59:56Z","cross_cats_sorted":[],"title_canon_sha256":"13c5fe8f86518ed5dfebc5925f4383c457aaeb1569aff80b58f99bec1f9a0bdd","abstract_canon_sha256":"911901d9dda25c21655e807eba7bca1d8c5ffd6af4d23544b1c9281c6698bd4a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:38:28.574083Z","signature_b64":"Vu34kepMRRWl0kK97vt9SFQzg4Y3bnidibKgDT2kaV33WbKocOGZv5m1QQ983iD/d34d9cNjDAecBE30JRUiBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"78e8c775e4b6d91f23cbb3ab5a827872f2a77c00bb2357e9c5c7461c7ac0bab5","last_reissued_at":"2026-07-05T10:38:28.573645Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:38:28.573645Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Tuning-Free Amodal Segmentation via the Occlusion-Free Bias of Inpainting Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bedrich Benes, Jae Joong Lee, Raymond A. Yeh","submitted_at":"2025-03-24T17:59:56Z","abstract_excerpt":"Amodal segmentation aims to predict segmentation masks for both the visible and occluded regions of an object. Most existing works formulate this as a supervised learning problem, requiring manually annotated amodal masks or synthetic training data. Consequently, their performance depends on the quality of the datasets, which often lack diversity and scale. This work introduces a tuning-free approach that repurposes pretrained diffusion-based inpainting models for amodal segmentation. Our approach is motivated by the \"occlusion-free bias\" of inpainting models, i.e., the inpainted objects tend "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.18947","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/2503.18947/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":"2503.18947","created_at":"2026-07-05T10:38:28.573702+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.18947v1","created_at":"2026-07-05T10:38:28.573702+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.18947","created_at":"2026-07-05T10:38:28.573702+00:00"},{"alias_kind":"pith_short_12","alias_value":"PDUMO5PEW3MR","created_at":"2026-07-05T10:38:28.573702+00:00"},{"alias_kind":"pith_short_16","alias_value":"PDUMO5PEW3MR6I6L","created_at":"2026-07-05T10:38:28.573702+00:00"},{"alias_kind":"pith_short_8","alias_value":"PDUMO5PE","created_at":"2026-07-05T10:38:28.573702+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.21544","citing_title":"DeOcc-1-to-3: 3D De-Occlusion from a Single Image via Self-Supervised Multi-View Diffusion","ref_index":36,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PDUMO5PEW3MR6I6LWOVVVATYOL","json":"https://pith.science/pith/PDUMO5PEW3MR6I6LWOVVVATYOL.json","graph_json":"https://pith.science/api/pith-number/PDUMO5PEW3MR6I6LWOVVVATYOL/graph.json","events_json":"https://pith.science/api/pith-number/PDUMO5PEW3MR6I6LWOVVVATYOL/events.json","paper":"https://pith.science/paper/PDUMO5PE"},"agent_actions":{"view_html":"https://pith.science/pith/PDUMO5PEW3MR6I6LWOVVVATYOL","download_json":"https://pith.science/pith/PDUMO5PEW3MR6I6LWOVVVATYOL.json","view_paper":"https://pith.science/paper/PDUMO5PE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.18947&json=true","fetch_graph":"https://pith.science/api/pith-number/PDUMO5PEW3MR6I6LWOVVVATYOL/graph.json","fetch_events":"https://pith.science/api/pith-number/PDUMO5PEW3MR6I6LWOVVVATYOL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PDUMO5PEW3MR6I6LWOVVVATYOL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PDUMO5PEW3MR6I6LWOVVVATYOL/action/storage_attestation","attest_author":"https://pith.science/pith/PDUMO5PEW3MR6I6LWOVVVATYOL/action/author_attestation","sign_citation":"https://pith.science/pith/PDUMO5PEW3MR6I6LWOVVVATYOL/action/citation_signature","submit_replication":"https://pith.science/pith/PDUMO5PEW3MR6I6LWOVVVATYOL/action/replication_record"}},"created_at":"2026-07-05T10:38:28.573702+00:00","updated_at":"2026-07-05T10:38:28.573702+00:00"}