{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:B5O7VR3N2F7DFVMTL4RHPELA4A","short_pith_number":"pith:B5O7VR3N","schema_version":"1.0","canonical_sha256":"0f5dfac76dd17e32d5935f22779160e007924f3730cdbee161198b83b39faf00","source":{"kind":"arxiv","id":"2005.02357","version":3},"attestation_state":"computed","paper":{"title":"Sub-Image Anomaly Detection with Deep Pyramid Correspondences","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Niv Cohen, Yedid Hoshen","submitted_at":"2020-05-05T17:43:35Z","abstract_excerpt":"Nearest neighbor (kNN) methods utilizing deep pre-trained features exhibit very strong anomaly detection performance when applied to entire images. A limitation of kNN methods is the lack of segmentation map describing where the anomaly lies inside the image. In this work we present a novel anomaly segmentation approach based on alignment between an anomalous image and a constant number of the similar normal images. Our method, Semantic Pyramid Anomaly Detection (SPADE) uses correspondences based on a multi-resolution feature pyramid. SPADE is shown to achieve state-of-the-art performance on u"},"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":"2005.02357","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2020-05-05T17:43:35Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"4aaa6a9d592f75c631329908398bb57d981e391045f5265f4742fc4857dfe838","abstract_canon_sha256":"fff83acd250de4ce3576c1a3e42b29b582cbd07e5ef730bfad0a506f161759c4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:12:29.400655Z","signature_b64":"YyJQTRvTdPF4RVeI6m5uSxprhT5XlUu32XSiRTWC4mtItK6SRKPNXg3CpgAn5IU3IrDLWsitZnvlC/4TQwKpDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0f5dfac76dd17e32d5935f22779160e007924f3730cdbee161198b83b39faf00","last_reissued_at":"2026-07-05T02:12:29.400208Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:12:29.400208Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Sub-Image Anomaly Detection with Deep Pyramid Correspondences","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Niv Cohen, Yedid Hoshen","submitted_at":"2020-05-05T17:43:35Z","abstract_excerpt":"Nearest neighbor (kNN) methods utilizing deep pre-trained features exhibit very strong anomaly detection performance when applied to entire images. A limitation of kNN methods is the lack of segmentation map describing where the anomaly lies inside the image. In this work we present a novel anomaly segmentation approach based on alignment between an anomalous image and a constant number of the similar normal images. Our method, Semantic Pyramid Anomaly Detection (SPADE) uses correspondences based on a multi-resolution feature pyramid. SPADE is shown to achieve state-of-the-art performance on u"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2005.02357","kind":"arxiv","version":3},"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/2005.02357/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":"2005.02357","created_at":"2026-07-05T02:12:29.400265+00:00"},{"alias_kind":"arxiv_version","alias_value":"2005.02357v3","created_at":"2026-07-05T02:12:29.400265+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2005.02357","created_at":"2026-07-05T02:12:29.400265+00:00"},{"alias_kind":"pith_short_12","alias_value":"B5O7VR3N2F7D","created_at":"2026-07-05T02:12:29.400265+00:00"},{"alias_kind":"pith_short_16","alias_value":"B5O7VR3N2F7DFVMT","created_at":"2026-07-05T02:12:29.400265+00:00"},{"alias_kind":"pith_short_8","alias_value":"B5O7VR3N","created_at":"2026-07-05T02:12:29.400265+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":10,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.25368","citing_title":"Hypergraph Normal World Models for Logical Visual Anomaly Detection","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2606.24375","citing_title":"MATCH: Flow Matching for Multi-View Anomaly Detection","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2606.23126","citing_title":"MambaADv2: Evolving Duality-enhanced State Space Model for Unsupervised Anomaly Detection","ref_index":45,"is_internal_anchor":false},{"citing_arxiv_id":"2606.29714","citing_title":"UniVAD v2: Unified Visual Anomaly Detection via Support-Conditioned Boundary Construction","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2605.28428","citing_title":"Anomaly as Non-Conformity via Training-Free Graph Laplacian Energy Minimization","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2605.29455","citing_title":"Uni-RCM: Unified Reference-guided Cross-modal Mapping for Multi-Class Anomaly Detection","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2605.23231","citing_title":"Beyond Normal References: Discriminative Few-Shot Anomaly Detection","ref_index":65,"is_internal_anchor":false},{"citing_arxiv_id":"2309.13904","citing_title":"Subspace-Guided Feature Reconstruction for Unsupervised Anomaly Localization","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2602.23013","citing_title":"SubspaceAD: Training-Free Few-Shot Anomaly Detection via Subspace Modeling","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05850","citing_title":"Align3D-AD: Cross-Modal Feature Alignment and Dual-Prompt Learning for Zero-shot 3D Anomaly Detection","ref_index":11,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/B5O7VR3N2F7DFVMTL4RHPELA4A","json":"https://pith.science/pith/B5O7VR3N2F7DFVMTL4RHPELA4A.json","graph_json":"https://pith.science/api/pith-number/B5O7VR3N2F7DFVMTL4RHPELA4A/graph.json","events_json":"https://pith.science/api/pith-number/B5O7VR3N2F7DFVMTL4RHPELA4A/events.json","paper":"https://pith.science/paper/B5O7VR3N"},"agent_actions":{"view_html":"https://pith.science/pith/B5O7VR3N2F7DFVMTL4RHPELA4A","download_json":"https://pith.science/pith/B5O7VR3N2F7DFVMTL4RHPELA4A.json","view_paper":"https://pith.science/paper/B5O7VR3N","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2005.02357&json=true","fetch_graph":"https://pith.science/api/pith-number/B5O7VR3N2F7DFVMTL4RHPELA4A/graph.json","fetch_events":"https://pith.science/api/pith-number/B5O7VR3N2F7DFVMTL4RHPELA4A/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/B5O7VR3N2F7DFVMTL4RHPELA4A/action/timestamp_anchor","attest_storage":"https://pith.science/pith/B5O7VR3N2F7DFVMTL4RHPELA4A/action/storage_attestation","attest_author":"https://pith.science/pith/B5O7VR3N2F7DFVMTL4RHPELA4A/action/author_attestation","sign_citation":"https://pith.science/pith/B5O7VR3N2F7DFVMTL4RHPELA4A/action/citation_signature","submit_replication":"https://pith.science/pith/B5O7VR3N2F7DFVMTL4RHPELA4A/action/replication_record"}},"created_at":"2026-07-05T02:12:29.400265+00:00","updated_at":"2026-07-05T02:12:29.400265+00:00"}