{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:B5QBWH5LEY37T3DQDHS6XN4EC6","short_pith_number":"pith:B5QBWH5L","schema_version":"1.0","canonical_sha256":"0f601b1fab2637f9ec7019e5ebb784178fece894448175a7eca05ece8cee973e","source":{"kind":"arxiv","id":"1906.03973","version":2},"attestation_state":"computed","paper":{"title":"E-LPIPS: Robust Perceptual Image Similarity via Random Transformation Ensembles","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.NE"],"primary_cat":"cs.CV","authors_text":"Erik H\\\"ark\\\"onen, Jaakko Lehtinen, Markus Kettunen","submitted_at":"2019-06-10T13:40:37Z","abstract_excerpt":"It has been recently shown that the hidden variables of convolutional neural networks make for an efficient perceptual similarity metric that accurately predicts human judgment on relative image similarity assessment. First, we show that such learned perceptual similarity metrics (LPIPS) are susceptible to adversarial attacks that dramatically contradict human visual similarity judgment. While this is not surprising in light of neural networks' well-known weakness to adversarial perturbations, we proceed to show that self-ensembling with an infinite family of random transformations of the inpu"},"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":"1906.03973","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2019-06-10T13:40:37Z","cross_cats_sorted":["cs.NE"],"title_canon_sha256":"4c870f7df2265464a465d7e7370fb66a1f829ec9e84fb0a2d92b08b0fef25249","abstract_canon_sha256":"ad0415e8100352d5b04e80c25966ad0a078d211816edbaf1de15935076e53311"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:43:42.076388Z","signature_b64":"I4zOSbqBSMcbYNrhtDoCMIxNDHy4dO1Kxh1sNpocTOUslbflBKJfQsMo1fQqPvVPeyXqS6j8wbB95vPNtK7WDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0f601b1fab2637f9ec7019e5ebb784178fece894448175a7eca05ece8cee973e","last_reissued_at":"2026-05-17T23:43:42.075569Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:43:42.075569Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"E-LPIPS: Robust Perceptual Image Similarity via Random Transformation Ensembles","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.NE"],"primary_cat":"cs.CV","authors_text":"Erik H\\\"ark\\\"onen, Jaakko Lehtinen, Markus Kettunen","submitted_at":"2019-06-10T13:40:37Z","abstract_excerpt":"It has been recently shown that the hidden variables of convolutional neural networks make for an efficient perceptual similarity metric that accurately predicts human judgment on relative image similarity assessment. First, we show that such learned perceptual similarity metrics (LPIPS) are susceptible to adversarial attacks that dramatically contradict human visual similarity judgment. While this is not surprising in light of neural networks' well-known weakness to adversarial perturbations, we proceed to show that self-ensembling with an infinite family of random transformations of the inpu"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1906.03973","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":""},"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":"1906.03973","created_at":"2026-05-17T23:43:42.075699+00:00"},{"alias_kind":"arxiv_version","alias_value":"1906.03973v2","created_at":"2026-05-17T23:43:42.075699+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1906.03973","created_at":"2026-05-17T23:43:42.075699+00:00"},{"alias_kind":"pith_short_12","alias_value":"B5QBWH5LEY37","created_at":"2026-05-18T12:33:12.712433+00:00"},{"alias_kind":"pith_short_16","alias_value":"B5QBWH5LEY37T3DQ","created_at":"2026-05-18T12:33:12.712433+00:00"},{"alias_kind":"pith_short_8","alias_value":"B5QBWH5L","created_at":"2026-05-18T12:33:12.712433+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"1907.05587","citing_title":"Stateful Detection of Black-Box Adversarial Attacks","ref_index":25,"is_internal_anchor":true},{"citing_arxiv_id":"2605.17500","citing_title":"The Silent Brush: Evaluating Artistic Style Leakage in AI Art Generation","ref_index":18,"is_internal_anchor":true},{"citing_arxiv_id":"2605.01718","citing_title":"Dual-branch Robust Unlearnable Examples","ref_index":6,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/B5QBWH5LEY37T3DQDHS6XN4EC6","json":"https://pith.science/pith/B5QBWH5LEY37T3DQDHS6XN4EC6.json","graph_json":"https://pith.science/api/pith-number/B5QBWH5LEY37T3DQDHS6XN4EC6/graph.json","events_json":"https://pith.science/api/pith-number/B5QBWH5LEY37T3DQDHS6XN4EC6/events.json","paper":"https://pith.science/paper/B5QBWH5L"},"agent_actions":{"view_html":"https://pith.science/pith/B5QBWH5LEY37T3DQDHS6XN4EC6","download_json":"https://pith.science/pith/B5QBWH5LEY37T3DQDHS6XN4EC6.json","view_paper":"https://pith.science/paper/B5QBWH5L","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1906.03973&json=true","fetch_graph":"https://pith.science/api/pith-number/B5QBWH5LEY37T3DQDHS6XN4EC6/graph.json","fetch_events":"https://pith.science/api/pith-number/B5QBWH5LEY37T3DQDHS6XN4EC6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/B5QBWH5LEY37T3DQDHS6XN4EC6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/B5QBWH5LEY37T3DQDHS6XN4EC6/action/storage_attestation","attest_author":"https://pith.science/pith/B5QBWH5LEY37T3DQDHS6XN4EC6/action/author_attestation","sign_citation":"https://pith.science/pith/B5QBWH5LEY37T3DQDHS6XN4EC6/action/citation_signature","submit_replication":"https://pith.science/pith/B5QBWH5LEY37T3DQDHS6XN4EC6/action/replication_record"}},"created_at":"2026-05-17T23:43:42.075699+00:00","updated_at":"2026-05-17T23:43:42.075699+00:00"}