{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:6NIQKDSFOSW3JHRJF32XEJI5US","short_pith_number":"pith:6NIQKDSF","schema_version":"1.0","canonical_sha256":"f351050e4574adb49e292ef572251da4bb3724265d4f0ab70d8e158c903360df","source":{"kind":"arxiv","id":"2604.05689","version":1},"attestation_state":"computed","paper":{"title":"CRFT: Consistent-Recurrent Feature Flow Transformer for Cross-Modal Image Registration","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"CRFT uses a transformer to learn a consistent recurrent feature flow that aligns cross-modal images more accurately and robustly than existing methods.","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Mengzhu Ding, Xichao Teng, Xuecong Liu, Zhang Li, Zixuan Sun","submitted_at":"2026-04-07T10:40:27Z","abstract_excerpt":"We present Consistent-Recurrent Feature Flow Transformer (CRFT), a unified coarse-to-fine framework based on feature flow learning for robust cross-modal image registration. CRFT learns a modality-independent feature flow representation within a transformer-based architecture that jointly performs feature alignment and flow estimation. The coarse stage establishes global correspondences through multi-scale feature correlation, while the fine stage refines local details via hierarchical feature fusion and adaptive spatial reasoning. To enhance geometric adaptability, an iterative discrepancy-gu"},"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":true,"formal_links_present":true},"canonical_record":{"source":{"id":"2604.05689","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-04-07T10:40:27Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"cc48b7e54d393b0777d6f02aa9d23aa590570315da226405be64278dd018b994","abstract_canon_sha256":"09cd875eb0e73572935920e3e662d2ea29577bdd4cac5262791b25635e48cad7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-30T02:18:08.151802Z","signature_b64":"yp8wdcS0qPzg+SybfDul890Bkfm7cYSnBpV4DGy80XLycpVC+AW9W6AsP7iCKL8eNF1pMAVa60AnwQkhCcqPBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f351050e4574adb49e292ef572251da4bb3724265d4f0ab70d8e158c903360df","last_reissued_at":"2026-06-30T02:18:08.151128Z","signature_status":"signed_v1","first_computed_at":"2026-06-30T02:18:08.151128Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CRFT: Consistent-Recurrent Feature Flow Transformer for Cross-Modal Image Registration","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"CRFT uses a transformer to learn a consistent recurrent feature flow that aligns cross-modal images more accurately and robustly than existing methods.","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Mengzhu Ding, Xichao Teng, Xuecong Liu, Zhang Li, Zixuan Sun","submitted_at":"2026-04-07T10:40:27Z","abstract_excerpt":"We present Consistent-Recurrent Feature Flow Transformer (CRFT), a unified coarse-to-fine framework based on feature flow learning for robust cross-modal image registration. CRFT learns a modality-independent feature flow representation within a transformer-based architecture that jointly performs feature alignment and flow estimation. The coarse stage establishes global correspondences through multi-scale feature correlation, while the fine stage refines local details via hierarchical feature fusion and adaptive spatial reasoning. To enhance geometric adaptability, an iterative discrepancy-gu"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"CRFT consistently outperforms state-of-the-art registration methods in both accuracy and robustness.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That a single modality-independent feature flow representation learned in a transformer can jointly handle feature alignment and flow estimation while the iterative discrepancy-guided attention with Spatial Geometric Transform enforces consistency under large affine and scale variations.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"CRFT is a new transformer architecture using recurrent consistent feature flow learning to achieve accurate and robust cross-modal image registration under large variations.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"CRFT uses a transformer to learn a consistent recurrent feature flow that aligns cross-modal images more accurately and robustly than existing methods.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"042c5fcdb45fe935cdcb410e49d6c1fa51c1e4e6d64cf100da03bbdbb48c5a1c"},"source":{"id":"2604.05689","kind":"arxiv","version":1},"verdict":{"id":"49fafc63-3361-4266-af2e-93324b8ef09a","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-10T19:34:16.218576Z","strongest_claim":"CRFT consistently outperforms state-of-the-art registration methods in both accuracy and robustness.","one_line_summary":"CRFT is a new transformer architecture using recurrent consistent feature flow learning to achieve accurate and robust cross-modal image registration under large variations.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That a single modality-independent feature flow representation learned in a transformer can jointly handle feature alignment and flow estimation while the iterative discrepancy-guided attention with Spatial Geometric Transform enforces consistency under large affine and scale variations.","pith_extraction_headline":"CRFT uses a transformer to learn a consistent recurrent feature flow that aligns cross-modal images more accurately and robustly than existing methods."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.05689/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":66,"sample":[{"doi":"","year":2024,"title":"Deep learning models for digital image processing: a review.Artificial Intelligence Review, 57(1):11","work_id":"10f36277-2b3b-4126-8c6e-27acf1e62bfc","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2025,"title":"Graphi2p: Image-to-point cloud registration with exploring pattern of correspondence via graph learning","work_id":"f9c7a4d2-7191-4470-bf62-3c05728a2b15","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"A survey on deep learning in medical image registration: New technologies, uncertainty, evaluation metrics, and beyond.Medical Image Analysis, 100:103385","work_id":"393cbdbe-f3a8-4461-a62b-14adea99f0be","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2024,"title":"Mvsplat: Efficient 3d gaussian splatting from sparse multi-view images","work_id":"7e9dc93b-bac5-4ee5-8ee6-9917a40c6c06","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2024,"title":"Dsap: Dynamic sparse attention perception matcher for accurate local feature matching","work_id":"3fe403b5-98e3-425e-b483-a3fe94790fa8","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":66,"snapshot_sha256":"6f6fd9388ae90e1b7e52e7aceef8ccf80ada3a918399797e2c585de13637f91d","internal_anchors":0},"formal_canon":{"evidence_count":2,"snapshot_sha256":"424e65ad8df31fda6f16ad8c9beede6a169ea1514d1e782905ad5cab8f1f8ae5"},"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":"2604.05689","created_at":"2026-06-30T02:18:08.151216+00:00"},{"alias_kind":"arxiv_version","alias_value":"2604.05689v1","created_at":"2026-06-30T02:18:08.151216+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.05689","created_at":"2026-06-30T02:18:08.151216+00:00"},{"alias_kind":"pith_short_12","alias_value":"6NIQKDSFOSW3","created_at":"2026-06-30T02:18:08.151216+00:00"},{"alias_kind":"pith_short_16","alias_value":"6NIQKDSFOSW3JHRJ","created_at":"2026-06-30T02:18:08.151216+00:00"},{"alias_kind":"pith_short_8","alias_value":"6NIQKDSF","created_at":"2026-06-30T02:18:08.151216+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":2,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6NIQKDSFOSW3JHRJF32XEJI5US","json":"https://pith.science/pith/6NIQKDSFOSW3JHRJF32XEJI5US.json","graph_json":"https://pith.science/api/pith-number/6NIQKDSFOSW3JHRJF32XEJI5US/graph.json","events_json":"https://pith.science/api/pith-number/6NIQKDSFOSW3JHRJF32XEJI5US/events.json","paper":"https://pith.science/paper/6NIQKDSF"},"agent_actions":{"view_html":"https://pith.science/pith/6NIQKDSFOSW3JHRJF32XEJI5US","download_json":"https://pith.science/pith/6NIQKDSFOSW3JHRJF32XEJI5US.json","view_paper":"https://pith.science/paper/6NIQKDSF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2604.05689&json=true","fetch_graph":"https://pith.science/api/pith-number/6NIQKDSFOSW3JHRJF32XEJI5US/graph.json","fetch_events":"https://pith.science/api/pith-number/6NIQKDSFOSW3JHRJF32XEJI5US/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6NIQKDSFOSW3JHRJF32XEJI5US/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6NIQKDSFOSW3JHRJF32XEJI5US/action/storage_attestation","attest_author":"https://pith.science/pith/6NIQKDSFOSW3JHRJF32XEJI5US/action/author_attestation","sign_citation":"https://pith.science/pith/6NIQKDSFOSW3JHRJF32XEJI5US/action/citation_signature","submit_replication":"https://pith.science/pith/6NIQKDSFOSW3JHRJF32XEJI5US/action/replication_record"}},"created_at":"2026-06-30T02:18:08.151216+00:00","updated_at":"2026-06-30T02:18:08.151216+00:00"}