{"paper":{"title":"TopoHR: Hierarchical Centerline Representation for Cyclic Topology Reasoning in Driving Scenes with Point-to-Instance Relations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Cyclic interaction between centerline detection and topology reasoning, driven by hierarchical point-to-instance features, improves road layout understanding in driving scenes.","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bo Song, Erkang Cheng, Haibin Ling, Yifeng Bai, Zhirong Chen","submitted_at":"2026-04-27T07:13:33Z","abstract_excerpt":"Topology reasoning is crucial for autonomous driving. Current methods primarily focus on instance-level learning for centerline detection, followed by a sequential module for topology reasoning that relies on simplified MLP layers. Moreover, they often neglect the importance of \\textit{point-to-instance} (P2I) relationships in topology reasoning. To address these limitations, we present TopoHR (Topological Hierarchical Representation), a novel end-to-end framework that establishes cyclic interaction between centerline detection and topology reasoning, allowing them to iteratively enhance each "},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"On the OpenLane-V2 benchmark, TopoHR refreshes state-of-the-art performance with significant improvements. Notably, compared with previous best results, TopoHR achieves +3.8 in DET_l, +5.4 in TOP_ll on subset_A and +11.0 in DET_l, +7.9 in TOP_ll on subset_B, validating the effectiveness of the proposed components.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the cyclic interaction between centerline detection and topology reasoning, enabled by the hierarchical point-to-instance features, is the primary driver of the reported benchmark gains rather than other unstated factors such as training details or dataset specifics.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"TopoHR proposes a hierarchical centerline representation and topology reasoning module with point-to-instance relations and cyclic interactions, achieving new state-of-the-art results on the OpenLane-V2 benchmark for autonomous driving.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Cyclic interaction between centerline detection and topology reasoning, driven by hierarchical point-to-instance features, improves road layout understanding in driving scenes.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"cd28373c42543c79bff1476ebec772bc3f2a35cfea5d9292d3d71783f69f0e2c"},"source":{"id":"2604.24119","kind":"arxiv","version":2},"verdict":{"id":"c53fd2ce-2a91-48bd-9ed8-3b6fa9da9f79","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-08T04:32:58.989154Z","strongest_claim":"On the OpenLane-V2 benchmark, TopoHR refreshes state-of-the-art performance with significant improvements. Notably, compared with previous best results, TopoHR achieves +3.8 in DET_l, +5.4 in TOP_ll on subset_A and +11.0 in DET_l, +7.9 in TOP_ll on subset_B, validating the effectiveness of the proposed components.","one_line_summary":"TopoHR proposes a hierarchical centerline representation and topology reasoning module with point-to-instance relations and cyclic interactions, achieving new state-of-the-art results on the OpenLane-V2 benchmark for autonomous driving.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the cyclic interaction between centerline detection and topology reasoning, enabled by the hierarchical point-to-instance features, is the primary driver of the reported benchmark gains rather than other unstated factors such as training details or dataset specifics.","pith_extraction_headline":"Cyclic interaction between centerline detection and topology reasoning, driven by hierarchical point-to-instance features, improves road layout understanding in driving scenes."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.24119/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"ai_meta_artifact","ran_at":"2026-05-21T07:38:38.138815Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T22:24:22.466179Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"7c4a7e82afbea55dc3b39f5de6605085b94fe6cf4088a7df0d4c30f012bc4222"},"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"}