{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:PGAL6F6CSHQQ7MLEE5S3QGXDJQ","short_pith_number":"pith:PGAL6F6C","schema_version":"1.0","canonical_sha256":"7980bf17c291e10fb1642765b81ae34c00d46de44c515b0c88a471f804a9c1dc","source":{"kind":"arxiv","id":"2103.17263","version":5},"attestation_state":"computed","paper":{"title":"Rethinking Self-supervised Correspondence Learning: A Video Frame-level Similarity Perspective","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Jiarui Xu, Xiaolong Wang","submitted_at":"2021-03-31T17:56:35Z","abstract_excerpt":"Learning a good representation for space-time correspondence is the key for various computer vision tasks, including tracking object bounding boxes and performing video object pixel segmentation. To learn generalizable representation for correspondence in large-scale, a variety of self-supervised pretext tasks are proposed to explicitly perform object-level or patch-level similarity learning. Instead of following the previous literature, we propose to learn correspondence using Video Frame-level Similarity (VFS) learning, i.e, simply learning from comparing video frames. Our work is inspired b"},"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":"2103.17263","kind":"arxiv","version":5},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2021-03-31T17:56:35Z","cross_cats_sorted":[],"title_canon_sha256":"54989ba1763dd5385ef3213e03299218dd00ffe899dd5ce6329315db9e66bbf1","abstract_canon_sha256":"2ab7c7348e24c5b4b1c7d8feb5a2b6b9993aeb75c6070a41d88a95e130a5d6b0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:22:34.455879Z","signature_b64":"fOLm8nVcfMfQ2Au7gv4owPrzyBcdy+WN9PV5sUwrMvH1U86EFp8re8J9gRnWEcsgUIk0uTPjouITH7aa1LLNAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7980bf17c291e10fb1642765b81ae34c00d46de44c515b0c88a471f804a9c1dc","last_reissued_at":"2026-07-05T03:22:34.455479Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:22:34.455479Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Rethinking Self-supervised Correspondence Learning: A Video Frame-level Similarity Perspective","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Jiarui Xu, Xiaolong Wang","submitted_at":"2021-03-31T17:56:35Z","abstract_excerpt":"Learning a good representation for space-time correspondence is the key for various computer vision tasks, including tracking object bounding boxes and performing video object pixel segmentation. To learn generalizable representation for correspondence in large-scale, a variety of self-supervised pretext tasks are proposed to explicitly perform object-level or patch-level similarity learning. Instead of following the previous literature, we propose to learn correspondence using Video Frame-level Similarity (VFS) learning, i.e, simply learning from comparing video frames. Our work is inspired b"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2103.17263","kind":"arxiv","version":5},"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/2103.17263/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":"2103.17263","created_at":"2026-07-05T03:22:34.455536+00:00"},{"alias_kind":"arxiv_version","alias_value":"2103.17263v5","created_at":"2026-07-05T03:22:34.455536+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2103.17263","created_at":"2026-07-05T03:22:34.455536+00:00"},{"alias_kind":"pith_short_12","alias_value":"PGAL6F6CSHQQ","created_at":"2026-07-05T03:22:34.455536+00:00"},{"alias_kind":"pith_short_16","alias_value":"PGAL6F6CSHQQ7MLE","created_at":"2026-07-05T03:22:34.455536+00:00"},{"alias_kind":"pith_short_8","alias_value":"PGAL6F6C","created_at":"2026-07-05T03:22:34.455536+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":55,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PGAL6F6CSHQQ7MLEE5S3QGXDJQ","json":"https://pith.science/pith/PGAL6F6CSHQQ7MLEE5S3QGXDJQ.json","graph_json":"https://pith.science/api/pith-number/PGAL6F6CSHQQ7MLEE5S3QGXDJQ/graph.json","events_json":"https://pith.science/api/pith-number/PGAL6F6CSHQQ7MLEE5S3QGXDJQ/events.json","paper":"https://pith.science/paper/PGAL6F6C"},"agent_actions":{"view_html":"https://pith.science/pith/PGAL6F6CSHQQ7MLEE5S3QGXDJQ","download_json":"https://pith.science/pith/PGAL6F6CSHQQ7MLEE5S3QGXDJQ.json","view_paper":"https://pith.science/paper/PGAL6F6C","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2103.17263&json=true","fetch_graph":"https://pith.science/api/pith-number/PGAL6F6CSHQQ7MLEE5S3QGXDJQ/graph.json","fetch_events":"https://pith.science/api/pith-number/PGAL6F6CSHQQ7MLEE5S3QGXDJQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PGAL6F6CSHQQ7MLEE5S3QGXDJQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PGAL6F6CSHQQ7MLEE5S3QGXDJQ/action/storage_attestation","attest_author":"https://pith.science/pith/PGAL6F6CSHQQ7MLEE5S3QGXDJQ/action/author_attestation","sign_citation":"https://pith.science/pith/PGAL6F6CSHQQ7MLEE5S3QGXDJQ/action/citation_signature","submit_replication":"https://pith.science/pith/PGAL6F6CSHQQ7MLEE5S3QGXDJQ/action/replication_record"}},"created_at":"2026-07-05T03:22:34.455536+00:00","updated_at":"2026-07-05T03:22:34.455536+00:00"}