{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:GGQQAFK33MP4OYUKFU5YXDMMX3","short_pith_number":"pith:GGQQAFK3","schema_version":"1.0","canonical_sha256":"31a100155bdb1fc7628a2d3b8b8d8cbed3271f53b77ff9a48f8853a3f674ece6","source":{"kind":"arxiv","id":"2311.14665","version":1},"attestation_state":"computed","paper":{"title":"Understanding Self-Supervised Features for Learning Unsupervised Instance Segmentation","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Christian Rupprecht, Iro Laina, Luke Melas-Kyriazi, Paul Engstler","submitted_at":"2023-11-24T18:55:53Z","abstract_excerpt":"Self-supervised learning (SSL) can be used to solve complex visual tasks without human labels. Self-supervised representations encode useful semantic information about images, and as a result, they have already been used for tasks such as unsupervised semantic segmentation. In this paper, we investigate self-supervised representations for instance segmentation without any manual annotations. We find that the features of different SSL methods vary in their level of instance-awareness. In particular, DINO features, which are known to be excellent semantic descriptors, lack behind MAE features in"},"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":"2311.14665","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2023-11-24T18:55:53Z","cross_cats_sorted":[],"title_canon_sha256":"742d5f649324f17b856b3cd06b014200fd0208aa6ff54df9a0a5d41a7315ff25","abstract_canon_sha256":"ac33965835387dc8b6135618b96da699e2889fc63be3876cc0e6f0a1c624cb83"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:16:22.484547Z","signature_b64":"TKXp1z1T2QDisufRBV9pVtSk4iB/2y9E7E78EqvsDhbjvWIOnwjmFGNEmlphKyacQ/lqcR6eW8oUzaZdAefNDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"31a100155bdb1fc7628a2d3b8b8d8cbed3271f53b77ff9a48f8853a3f674ece6","last_reissued_at":"2026-07-05T07:16:22.484029Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:16:22.484029Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Understanding Self-Supervised Features for Learning Unsupervised Instance Segmentation","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Christian Rupprecht, Iro Laina, Luke Melas-Kyriazi, Paul Engstler","submitted_at":"2023-11-24T18:55:53Z","abstract_excerpt":"Self-supervised learning (SSL) can be used to solve complex visual tasks without human labels. Self-supervised representations encode useful semantic information about images, and as a result, they have already been used for tasks such as unsupervised semantic segmentation. In this paper, we investigate self-supervised representations for instance segmentation without any manual annotations. We find that the features of different SSL methods vary in their level of instance-awareness. In particular, DINO features, which are known to be excellent semantic descriptors, lack behind MAE features in"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.14665","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/2311.14665/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":"2311.14665","created_at":"2026-07-05T07:16:22.484083+00:00"},{"alias_kind":"arxiv_version","alias_value":"2311.14665v1","created_at":"2026-07-05T07:16:22.484083+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.14665","created_at":"2026-07-05T07:16:22.484083+00:00"},{"alias_kind":"pith_short_12","alias_value":"GGQQAFK33MP4","created_at":"2026-07-05T07:16:22.484083+00:00"},{"alias_kind":"pith_short_16","alias_value":"GGQQAFK33MP4OYUK","created_at":"2026-07-05T07:16:22.484083+00:00"},{"alias_kind":"pith_short_8","alias_value":"GGQQAFK3","created_at":"2026-07-05T07:16:22.484083+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.18267","citing_title":"MARCO: Navigating the Unseen Space of Semantic Correspondence","ref_index":126,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GGQQAFK33MP4OYUKFU5YXDMMX3","json":"https://pith.science/pith/GGQQAFK33MP4OYUKFU5YXDMMX3.json","graph_json":"https://pith.science/api/pith-number/GGQQAFK33MP4OYUKFU5YXDMMX3/graph.json","events_json":"https://pith.science/api/pith-number/GGQQAFK33MP4OYUKFU5YXDMMX3/events.json","paper":"https://pith.science/paper/GGQQAFK3"},"agent_actions":{"view_html":"https://pith.science/pith/GGQQAFK33MP4OYUKFU5YXDMMX3","download_json":"https://pith.science/pith/GGQQAFK33MP4OYUKFU5YXDMMX3.json","view_paper":"https://pith.science/paper/GGQQAFK3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2311.14665&json=true","fetch_graph":"https://pith.science/api/pith-number/GGQQAFK33MP4OYUKFU5YXDMMX3/graph.json","fetch_events":"https://pith.science/api/pith-number/GGQQAFK33MP4OYUKFU5YXDMMX3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GGQQAFK33MP4OYUKFU5YXDMMX3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GGQQAFK33MP4OYUKFU5YXDMMX3/action/storage_attestation","attest_author":"https://pith.science/pith/GGQQAFK33MP4OYUKFU5YXDMMX3/action/author_attestation","sign_citation":"https://pith.science/pith/GGQQAFK33MP4OYUKFU5YXDMMX3/action/citation_signature","submit_replication":"https://pith.science/pith/GGQQAFK33MP4OYUKFU5YXDMMX3/action/replication_record"}},"created_at":"2026-07-05T07:16:22.484083+00:00","updated_at":"2026-07-05T07:16:22.484083+00:00"}