{"url":"/sota/point-cloud-registration-on-3dmatch-at-least-2","task":{"name":"Point Cloud Registration","url":"/task/point-cloud-registration","note":null},"dataset":{"name":"3DMatch (at least 30% overlapped - sample 5k interest points)","url":null},"category":"Computer Vision","categories":["Computer Vision"],"category_note":null,"description":"**Point Cloud Registration** is a fundamental problem in 3D computer vision and photogrammetry. Given several sets of points in different coordinate systems, the aim of registration is to find the transformation that best aligns all of them into a common coordinate system. Point Cloud Registration plays a significant role in many vision applications such as 3D model reconstruction, cultural heritage management, landslide monitoring and solar energy analysis.\r\n\r\n\r\n<span class=\"description-source\">Source: [Iterative Global Similarity Points : A robust coarse-to-fine integration solution for pairwise 3D point cloud registration ](https://arxiv.org/abs/1808.03899)</span>","description_from":"task","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","rank":"the archive's row order at snapshot; not re-ranked","rows_end_at":"2025-07-28","rows_withheld_as_spam":0,"metric_values":"the archive's strings, untouched"},"metrics":["Recall ( correspondence RMSE below 0.2)"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"Recall ( correspondence RMSE below 0.2)":null}},"counts":{"rows":11,"rows_with_code":11,"rows_with_paper_page":11,"rows_dated":11,"rows_using_additional_data":0},"rows":[{"rank_in_archive_order":1,"model":"NgeNet","metrics":{"Recall ( correspondence RMSE below 0.2)":"92.9"},"uses_additional_data":false,"paper_date":"2022-01-28","paper":"/paper/neighborhood-aware-geometric-encoding-network","paper_url":"https://arxiv.org/abs/2201.12094v2","paper_title":"Leveraging Inlier Correspondences Proportion for Point Cloud Registration","code":"https://github.com/zhulf0804/ngenet","n_code_links":1,"syntology":null},{"rank_in_archive_order":2,"model":"REGTR","metrics":{"Recall ( correspondence RMSE below 0.2)":"92"},"uses_additional_data":false,"paper_date":"2022-03-28","paper":"/paper/regtr-end-to-end-point-cloud-correspondences","paper_url":"https://arxiv.org/abs/2203.14517v1","paper_title":"REGTR: End-to-end Point Cloud Correspondences with Transformers","code":"https://github.com/yewzijian/regtr","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":16,"n_samples":16,"n_pointer_only_licence":0}},{"rank_in_archive_order":3,"model":"Predator-1k","metrics":{"Recall ( correspondence RMSE below 0.2)":"90.5"},"uses_additional_data":false,"paper_date":"2020-11-25","paper":"/paper/predator-registration-of-3d-point-clouds-with","paper_url":"https://arxiv.org/abs/2011.13005v3","paper_title":"PREDATOR: Registration of 3D Point Clouds with Low Overlap","code":"https://github.com/qinzheng93/geotransformer","n_code_links":5,"syntology":{"n_ran":0,"n_unverified":1,"n_samples":1,"n_pointer_only_licence":1}},{"rank_in_archive_order":4,"model":"Predator-5k","metrics":{"Recall ( correspondence RMSE below 0.2)":"89"},"uses_additional_data":false,"paper_date":"2020-11-25","paper":"/paper/predator-registration-of-3d-point-clouds-with","paper_url":"https://arxiv.org/abs/2011.13005v3","paper_title":"PREDATOR: Registration of 3D Point Clouds with Low Overlap","code":"https://github.com/qinzheng93/geotransformer","n_code_links":5,"syntology":{"n_ran":0,"n_unverified":1,"n_samples":1,"n_pointer_only_licence":1}},{"rank_in_archive_order":5,"model":"PCAM (reported in REGTR)","metrics":{"Recall ( correspondence RMSE below 0.2)":"85.5"},"uses_additional_data":false,"paper_date":"2021-10-04","paper":"/paper/pcam-product-of-cross-attention-matrices-for","paper_url":"https://arxiv.org/abs/2110.01269v1","paper_title":"PCAM: Product of Cross-Attention Matrices for Rigid Registration of Point Clouds","code":"https://github.com/valeoai/pcam","n_code_links":1,"syntology":null},{"rank_in_archive_order":6,"model":"DGR (reported in REGTR)","metrics":{"Recall ( correspondence RMSE below 0.2)":"85.3"},"uses_additional_data":false,"paper_date":"2020-04-24","paper":"/paper/deep-global-registration","paper_url":"https://arxiv.org/abs/2004.11540v2","paper_title":"Deep Global Registration","code":"https://github.com/chrischoy/FCGF","n_code_links":2,"syntology":{"n_ran":5,"n_unverified":4,"n_samples":9,"n_pointer_only_licence":0}},{"rank_in_archive_order":7,"model":"FCGF (reported in PREDATOR)","metrics":{"Recall ( correspondence RMSE below 0.2)":"85.1"},"uses_additional_data":false,"paper_date":"2019-10-27","paper":"/paper/fully-convolutional-geometric-features","paper_url":"https://github.com/chrischoy/FCGF","paper_title":"Fully Convolutional Geometric Features","code":"https://github.com/chrischoy/FCGF","n_code_links":1,"syntology":null},{"rank_in_archive_order":8,"model":"D3Feat (reported in PREDATOR)","metrics":{"Recall ( correspondence RMSE below 0.2)":"81.6"},"uses_additional_data":false,"paper_date":"2020-03-06","paper":"/paper/d3feat-joint-learning-of-dense-detection-and","paper_url":"https://arxiv.org/abs/2003.03164v1","paper_title":"D3Feat: Joint Learning of Dense Detection and Description of 3D Local Features","code":"https://github.com/XuyangBai/D3Feat","n_code_links":2,"syntology":{"n_ran":0,"n_unverified":8,"n_samples":8,"n_pointer_only_licence":0}},{"rank_in_archive_order":9,"model":"3DSN (reported in PREDATOR)","metrics":{"Recall ( correspondence RMSE below 0.2)":"78.4"},"uses_additional_data":false,"paper_date":"2018-11-16","paper":"/paper/the-perfect-match-3d-point-cloud-matching","paper_url":"https://arxiv.org/abs/1811.06879v3","paper_title":"The Perfect Match: 3D Point Cloud Matching with Smoothed Densities","code":"https://github.com/zgojcic/3DSmoothNet","n_code_links":2,"syntology":{"n_ran":1,"n_unverified":7,"n_samples":8,"n_pointer_only_licence":0}},{"rank_in_archive_order":10,"model":"Predator-NR","metrics":{"Recall ( correspondence RMSE below 0.2)":"62.7"},"uses_additional_data":false,"paper_date":"2020-11-25","paper":"/paper/predator-registration-of-3d-point-clouds-with","paper_url":"https://arxiv.org/abs/2011.13005v3","paper_title":"PREDATOR: Registration of 3D Point Clouds with Low Overlap","code":"https://github.com/qinzheng93/geotransformer","n_code_links":5,"syntology":{"n_ran":0,"n_unverified":1,"n_samples":1,"n_pointer_only_licence":1}},{"rank_in_archive_order":11,"model":"OMNet (reported in REGTR)","metrics":{"Recall ( correspondence RMSE below 0.2)":"35.9"},"uses_additional_data":false,"paper_date":"2021-03-01","paper":"/paper/omnet-learning-overlapping-mask-for-partial","paper_url":"https://arxiv.org/abs/2103.00937v5","paper_title":"OMNet: Learning Overlapping Mask for Partial-to-Partial Point Cloud Registration","code":"https://github.com/megvii-research/omnet","n_code_links":1,"syntology":null}],"since_archive":{"claim":"Results that newer papers report for their own method, placed here by Syntology. A model pointed at the cell in the paper's own table; the number was read from that cell and checked against this leaderboard's metric, dataset, split and scale; an independent check that saw this leaderboard's other rows and every other leaderboard on the same dataset accepted it. Not reviewed by the paper's authors or by the archive's editors, and not ranked against the archive rows.","extraction_file_present":true,"measurement":{"test_papers":883,"papers_with_output":881,"judged_true":108,"judged":110,"wilson95_lower":0.9361,"measured_on":"2026-09-24","frozen_commit":"0e3de0df94"},"measurement_note":"blind adjudication of accepted entries on a held-out split of archive papers, rules frozen before the test","coverage":{"sentence":"Syntology has checked 6,264 of the 9,581 papers on this site that are newer than the archive; results from the others appear after they are checked.","complete":false,"papers_newer_than_archive":9581,"papers_checked":6264,"papers_extracted_not_yet_verified":0,"boards_without_verdict":2,"papers_not_yet_extracted":3316},"order":"newest first by month (arXiv date, else the arXiv-id month), then arXiv id descending","columns":[],"entries":[]},"syntology":{"read_at":"2026-09-24T18:15:14+00:00","claim":"Per row: N of M harvested code samples from that row's paper executed on a synthesized fixture; the other M-N are unverified. Not a reproduction of the row's number; not a correctness claim. n_pointer_only_licence counts samples the site points at rather than redistributes (a licence axis, independent of ran/unverified).","rows_with_graph_line":7,"rows_with_any_sample_ran":2,"distinct_papers_with_graph_line":5,"distinct_papers_with_any_sample_ran":2,"samples_over_distinct_papers":{"n_ran":6,"n_unverified":36,"n_samples":42,"n_pointer_only_licence":1,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":6,"n_unverified":38,"n_samples":44,"n_pointer_only_licence":3,"note":"row-weighted: a paper behind several rows is counted once per row; inflated relative to samples_over_distinct_papers by design, kept for readers summing the per-row syntology blocks"}}}