{"paper":{"title":"L-PCN: A Point Cloud Accelerator Exploiting Spatial Locality through Octree-based Islandization","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"L-PCN partitions point clouds into octree islands to reuse overlapping subset data and cut repetitive feature operations in PCNs.","cross_cats":[],"primary_cat":"cs.AR","authors_text":"Bowen Jiang, Herman Lam, Jieming Yin, Jiliang Zhang, Xiangru Chen, Yiming Gao, Yuxiang Wang, Zhilei Chai","submitted_at":"2026-04-12T16:24:14Z","abstract_excerpt":"Existing Point Cloud Networks (PCNs) have proven to achieve great success in many point cloud tasks such as object part segmentation, shape classification, and so on. The most popular point-based PCNs are usually composed of two sequential steps: Data Structuring (DS) and Feature Computation (FC). In this paper, we first describe an important characteristic of the PCN-specific DS step that has not been addressed in existing PCN accelerators: the spatial locality resulting from overlapping points of the gathered point subsets. Using algorithm-hardware co-design, L-PCN (Locality-aware PCN) propo"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"L-PCN achieves a theoretical reduction in feature fetching ranging from 55.2% to 93.8% and in feature computation ranging from 45.4% to 80.6% during the PCN process. For experimentation, prototype L-PCN accelerators are implemented on the Intel Arria 10 GX FPGA. Experimental results prove that with the Islandization Unit as a plug-in, state-of-the-art PCN accelerators can achieve an additional speedup ranging from 1.2x to 3.2x.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The assumption that the spatial locality arising from overlapping point subsets in the data-structuring step is both substantial and stable enough across typical point-cloud workloads that the added partitioning and scheduling overhead does not offset the reported savings.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"L-PCN exploits spatial locality in point cloud networks via octree partitioning into islands and intra-island hub scheduling, delivering 55-94% less feature fetching, 45-81% less computation, and 1.2-3.2x additional speedup on FPGA prototypes.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"L-PCN partitions point clouds into octree islands to reuse overlapping subset data and cut repetitive feature operations in PCNs.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"ee3587b2a51ebb7571a200453d4ca5366a7fd2f9d53dd71453be6fbf59c81700"},"source":{"id":"2604.10716","kind":"arxiv","version":3},"verdict":{"id":"d80b0315-e05c-443a-8c6a-196a4f10a1d7","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-10T15:24:32.734362Z","strongest_claim":"L-PCN achieves a theoretical reduction in feature fetching ranging from 55.2% to 93.8% and in feature computation ranging from 45.4% to 80.6% during the PCN process. For experimentation, prototype L-PCN accelerators are implemented on the Intel Arria 10 GX FPGA. Experimental results prove that with the Islandization Unit as a plug-in, state-of-the-art PCN accelerators can achieve an additional speedup ranging from 1.2x to 3.2x.","one_line_summary":"L-PCN exploits spatial locality in point cloud networks via octree partitioning into islands and intra-island hub scheduling, delivering 55-94% less feature fetching, 45-81% less computation, and 1.2-3.2x additional speedup on FPGA prototypes.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The assumption that the spatial locality arising from overlapping point subsets in the data-structuring step is both substantial and stable enough across typical point-cloud workloads that the added partitioning and scheduling overhead does not offset the reported savings.","pith_extraction_headline":"L-PCN partitions point clouds into octree islands to reuse overlapping subset data and cut repetitive feature operations in PCNs."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.10716/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"}