{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:VO6AYZHVBSATHP6JJNYBVN4QJR","short_pith_number":"pith:VO6AYZHV","schema_version":"1.0","canonical_sha256":"abbc0c64f50c8133bfc94b701ab7904c51bf4380ee497e4cd20dede52edb20d8","source":{"kind":"arxiv","id":"2402.02352","version":4},"attestation_state":"computed","paper":{"title":"Region-Based Representations Revisited","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Ansel Blume, Derek Hoiem, Heyi Tao, Jae Yong Lee, Michal Shlapentokh-Rothman, Sethuraman T V, Wilfredo Torres, Yao Xiao, Yuqun Wu, Yu-Xiong Wang","submitted_at":"2024-02-04T05:33:04Z","abstract_excerpt":"We investigate whether region-based representations are effective for recognition. Regions were once a mainstay in recognition approaches, but pixel and patch-based features are now used almost exclusively. We show that recent class-agnostic segmenters like SAM can be effectively combined with strong unsupervised representations like DINOv2 and used for a wide variety of tasks, including semantic segmentation, object-based image retrieval, and multi-image analysis. Once the masks and features are extracted, these representations, even with linear decoders, enable competitive performance, makin"},"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":"2402.02352","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-02-04T05:33:04Z","cross_cats_sorted":[],"title_canon_sha256":"83af4bba48808d347ee0380c49c6e525e2f62d6c6ffe1b52f2fa3fbd2c35e23a","abstract_canon_sha256":"46d8ce500733b2db35b99ea9333d436b88f2318ff5fad9821d520983cf598f73"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:29:21.166290Z","signature_b64":"Ou38UItjzn/WcJ3VxrQzSD5dkIn07DOXGud5mtFxcHEXtaMQYHa4VIvTIrrAruSMYrv5nykV2cchxZX8B9KhDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"abbc0c64f50c8133bfc94b701ab7904c51bf4380ee497e4cd20dede52edb20d8","last_reissued_at":"2026-07-05T08:29:21.165723Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:29:21.165723Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Region-Based Representations Revisited","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Ansel Blume, Derek Hoiem, Heyi Tao, Jae Yong Lee, Michal Shlapentokh-Rothman, Sethuraman T V, Wilfredo Torres, Yao Xiao, Yuqun Wu, Yu-Xiong Wang","submitted_at":"2024-02-04T05:33:04Z","abstract_excerpt":"We investigate whether region-based representations are effective for recognition. Regions were once a mainstay in recognition approaches, but pixel and patch-based features are now used almost exclusively. We show that recent class-agnostic segmenters like SAM can be effectively combined with strong unsupervised representations like DINOv2 and used for a wide variety of tasks, including semantic segmentation, object-based image retrieval, and multi-image analysis. Once the masks and features are extracted, these representations, even with linear decoders, enable competitive performance, makin"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.02352","kind":"arxiv","version":4},"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/2402.02352/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":"2402.02352","created_at":"2026-07-05T08:29:21.165798+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.02352v4","created_at":"2026-07-05T08:29:21.165798+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.02352","created_at":"2026-07-05T08:29:21.165798+00:00"},{"alias_kind":"pith_short_12","alias_value":"VO6AYZHVBSAT","created_at":"2026-07-05T08:29:21.165798+00:00"},{"alias_kind":"pith_short_16","alias_value":"VO6AYZHVBSATHP6J","created_at":"2026-07-05T08:29:21.165798+00:00"},{"alias_kind":"pith_short_8","alias_value":"VO6AYZHV","created_at":"2026-07-05T08:29:21.165798+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.05543","citing_title":"FRAME: Pre-Training Video Feature Representations via Anticipation and Memory","ref_index":44,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VO6AYZHVBSATHP6JJNYBVN4QJR","json":"https://pith.science/pith/VO6AYZHVBSATHP6JJNYBVN4QJR.json","graph_json":"https://pith.science/api/pith-number/VO6AYZHVBSATHP6JJNYBVN4QJR/graph.json","events_json":"https://pith.science/api/pith-number/VO6AYZHVBSATHP6JJNYBVN4QJR/events.json","paper":"https://pith.science/paper/VO6AYZHV"},"agent_actions":{"view_html":"https://pith.science/pith/VO6AYZHVBSATHP6JJNYBVN4QJR","download_json":"https://pith.science/pith/VO6AYZHVBSATHP6JJNYBVN4QJR.json","view_paper":"https://pith.science/paper/VO6AYZHV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.02352&json=true","fetch_graph":"https://pith.science/api/pith-number/VO6AYZHVBSATHP6JJNYBVN4QJR/graph.json","fetch_events":"https://pith.science/api/pith-number/VO6AYZHVBSATHP6JJNYBVN4QJR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VO6AYZHVBSATHP6JJNYBVN4QJR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VO6AYZHVBSATHP6JJNYBVN4QJR/action/storage_attestation","attest_author":"https://pith.science/pith/VO6AYZHVBSATHP6JJNYBVN4QJR/action/author_attestation","sign_citation":"https://pith.science/pith/VO6AYZHVBSATHP6JJNYBVN4QJR/action/citation_signature","submit_replication":"https://pith.science/pith/VO6AYZHVBSATHP6JJNYBVN4QJR/action/replication_record"}},"created_at":"2026-07-05T08:29:21.165798+00:00","updated_at":"2026-07-05T08:29:21.165798+00:00"}