{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:SJSH7EMD4XIPTJ5USCEJ3R6SPD","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"720fa3578ff9b9c4880bf1757dd510bc21630ac8e1f78cddd893220cac73f52a","cross_cats_sorted":["cs.RO"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-03-03T23:11:17Z","title_canon_sha256":"3fbbb7823633c74ab00d462b5bb9d90efe5afece4d28661a1a164f37b92b3cdd"},"schema_version":"1.0","source":{"id":"2603.03577","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2603.03577","created_at":"2026-05-17T23:39:15Z"},{"alias_kind":"arxiv_version","alias_value":"2603.03577v2","created_at":"2026-05-17T23:39:15Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.03577","created_at":"2026-05-17T23:39:15Z"},{"alias_kind":"pith_short_12","alias_value":"SJSH7EMD4XIP","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"SJSH7EMD4XIPTJ5U","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"SJSH7EMD","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:7cf9d4272fb4b214dc405ace9d426a3d1310d25de4be08f16355623a912cacaf","target":"graph","created_at":"2026-05-17T23:39:15Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"L2G-Det bypasses explicit object proposals by leveraging dense patch-level matching between templates and the query image. Locally matched patches generate candidate points, which are refined through a candidate selection module to suppress false positives. The filtered points are then used to prompt an augmented Segment Anything Model (SAM) with instance-specific object tokens, enabling reliable reconstruction of complete instance masks."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That dense patch-level matching will produce sufficiently accurate candidate points in cluttered scenes and that the candidate selection module will reliably suppress false positives without removing true matches, allowing the augmented SAM to reconstruct complete masks."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"L2G-Det detects and segments novel object instances in open scenes by using local template patch matches to generate points that prompt an augmented SAM for global masks."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"L2G-Det detects and segments novel object instances in cluttered scenes by matching dense patches from templates to prompt an augmented SAM model."}],"snapshot_sha256":"2e142d49d42fab257bf1a5d9a03e2117a28a7ff07a016ddcc3c497f7b48f9923"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"paper":{"abstract_excerpt":"Detecting and segmenting novel object instances in open-world environments is a fundamental problem in robotic perception. Given only a small set of template images, a robot must locate and segment a specific object instance in a cluttered, previously unseen scene. Existing proposal-based approaches are highly sensitive to proposal quality and often fail under occlusion and background clutter. We propose L2G-Det, a local-to-global instance detection framework that bypasses explicit object proposals by leveraging dense patch-level matching between templates and the query image. Locally matched ","authors_text":"Jikai Wang, Qifan Zhang, Sai Haneesh Allu, Yangxiao Lu, Yu Xiang","cross_cats":["cs.RO"],"headline":"L2G-Det detects and segments novel object instances in cluttered scenes by matching dense patches from templates to prompt an augmented SAM model.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-03-03T23:11:17Z","title":"From Local Matches to Global Masks: Template-Guided Instance Detection and Segmentation in Open-World Scenes"},"references":{"count":51,"internal_anchors":10,"resolved_work":51,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"A modular robotic system for autonomous ex- ploration and semantic updating in large-scale indoor en- vironments, 2025","work_id":"e70e084b-1df0-41db-84ab-2733d8058757","year":2025},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"Target driven instance detection.arXiv preprint arXiv:1803.04610, 2018","work_id":"043d7cf2-4406-412c-b222-8326da0ceade","year":2018},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"Surf: Speeded up robust features","work_id":"ca56b6f8-4347-4b28-8acf-17fc693e7b73","year":2006},{"cited_arxiv_id":"2504.13181","doi":"","is_internal_anchor":true,"ref_index":4,"title":"Perception Encoder: The best visual embeddings are not at the output of the network","work_id":"409be941-4d4a-4ceb-a28a-eaa2d7709a1c","year":2025},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"Bidirectional attention network for monocular depth estimation","work_id":"fa6757b4-908a-45ef-8673-8cd2a926e54a","year":2021}],"snapshot_sha256":"3588ab9bb5c104589533b53de993d999394134d590c6f7a51898053ab0e6fa8b"},"source":{"id":"2603.03577","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-15T16:16:02.135381Z","id":"4486544c-a7f3-4f18-8381-f196e391ffdf","model_set":{"reader":"grok-4.3"},"one_line_summary":"L2G-Det detects and segments novel object instances in open scenes by using local template patch matches to generate points that prompt an augmented SAM for global masks.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"L2G-Det detects and segments novel object instances in cluttered scenes by matching dense patches from templates to prompt an augmented SAM model.","strongest_claim":"L2G-Det bypasses explicit object proposals by leveraging dense patch-level matching between templates and the query image. Locally matched patches generate candidate points, which are refined through a candidate selection module to suppress false positives. The filtered points are then used to prompt an augmented Segment Anything Model (SAM) with instance-specific object tokens, enabling reliable reconstruction of complete instance masks.","weakest_assumption":"That dense patch-level matching will produce sufficiently accurate candidate points in cluttered scenes and that the candidate selection module will reliably suppress false positives without removing true matches, allowing the augmented SAM to reconstruct complete masks."}},"verdict_id":"4486544c-a7f3-4f18-8381-f196e391ffdf"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:811fb05ef151d6f14befeccda859f1d3f7026e1c6cf2e14020ec0f70f8454091","target":"record","created_at":"2026-05-17T23:39:15Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"720fa3578ff9b9c4880bf1757dd510bc21630ac8e1f78cddd893220cac73f52a","cross_cats_sorted":["cs.RO"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-03-03T23:11:17Z","title_canon_sha256":"3fbbb7823633c74ab00d462b5bb9d90efe5afece4d28661a1a164f37b92b3cdd"},"schema_version":"1.0","source":{"id":"2603.03577","kind":"arxiv","version":2}},"canonical_sha256":"92647f9183e5d0f9a7b490889dc7d278d0acb87a8e872f6de0e1636e098290a8","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"92647f9183e5d0f9a7b490889dc7d278d0acb87a8e872f6de0e1636e098290a8","first_computed_at":"2026-05-17T23:39:15.873601Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:39:15.873601Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"fcc7buf69m0N47LVG4ahWeiHNSnlOXKnMKOWsjCQ7OBfuUrEIZ+yEBKS/d/x5gLe1N6N1fTdlxd/ZQus4Wu5Bg==","signature_status":"signed_v1","signed_at":"2026-05-17T23:39:15.874374Z","signed_message":"canonical_sha256_bytes"},"source_id":"2603.03577","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:811fb05ef151d6f14befeccda859f1d3f7026e1c6cf2e14020ec0f70f8454091","sha256:7cf9d4272fb4b214dc405ace9d426a3d1310d25de4be08f16355623a912cacaf"],"state_sha256":"7bcd588042b782b59f481202f4076c10fb7af14fae68e13b831383a87d3f9c5d"}