{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,21]],"date-time":"2026-02-21T18:37:23Z","timestamp":1771699043260,"version":"3.50.1"},"reference-count":21,"publisher":"MDPI AG","issue":"18","license":[{"start":{"date-parts":[[2020,9,7]],"date-time":"2020-09-07T00:00:00Z","timestamp":1599436800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100001809","name":"the National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["No. U1913602"],"award-info":[{"award-number":["No. U1913602"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"name":"the National Key R&amp;D Program of China","award":["No.2018YFB1305504"],"award-info":[{"award-number":["No.2018YFB1305504"]}]},{"name":"the Natural Science Basic Research Plan in Shaanxi Province of China","award":["No. 2020JQ-210"],"award-info":[{"award-number":["No. 2020JQ-210"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Sensors"],"abstract":"<jats:p>The local reference frame (LRF) acts as a critical role in 3D local shape description and matching. However, most existing LRFs are hand-crafted and suffer from limited repeatability and robustness. This paper presents the first attempt to learn an LRF via a Siamese network that needs weak supervision only. In particular, we argue that each neighboring point in the local surface gives a unique contribution to LRF construction and measure such contributions via learned weights. Extensive analysis and comparative experiments on three public datasets addressing different application scenarios have demonstrated that LRF-Net is more repeatable and robust than several state-of-the-art LRF methods (LRF-Net is only trained on one dataset). We show that LRFNet achieves 0.686 MeanCos performance on the UWA 3D modeling (UWA3M) dataset, outperforming the closest method by 0.18. In addition, LRF-Net can significantly boost the local shape description and 6-DoF pose estimation performance when matching 3D point clouds.<\/jats:p>","DOI":"10.3390\/s20185086","type":"journal-article","created":{"date-parts":[[2020,9,7]],"date-time":"2020-09-07T09:18:16Z","timestamp":1599470296000},"page":"5086","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":12,"title":["LRF-Net: Learning Local Reference Frames for 3D Local Shape Description and Matching"],"prefix":"10.3390","volume":"20","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-2787-4499","authenticated-orcid":false,"given":"Angfan","family":"Zhu","sequence":"first","affiliation":[{"name":"National Key Laboratory of Science and Technology on Multi-Spectral Information Processing, School of Artificial Intelligence and Automation, Huazhong University of Science and Technology, Wuhan 430074, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jiaqi","family":"Yang","sequence":"additional","affiliation":[{"name":"National Engineering Laboratory for Integrated aero-Space-Ground-Ocean Big Data Application Technology, School of Computer Science, Northwestern Polytechnical University, Xi\u2019an 710129, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Weiyue","family":"Zhao","sequence":"additional","affiliation":[{"name":"National Key Laboratory of Science and Technology on Multi-Spectral Information Processing, School of Artificial Intelligence and Automation, Huazhong University of Science and Technology, Wuhan 430074, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9223-1863","authenticated-orcid":false,"given":"Zhiguo","family":"Cao","sequence":"additional","affiliation":[{"name":"National Key Laboratory of Science and Technology on Multi-Spectral Information Processing, School of Artificial Intelligence and Automation, Huazhong University of Science and Technology, Wuhan 430074, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1968","published-online":{"date-parts":[[2020,9,7]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"63","DOI":"10.1007\/s11263-013-0627-y","article-title":"Rotational projection statistics for 3D local surface description and object recognition","volume":"105","author":"Guo","year":"2013","journal-title":"Int. J. Comput. Vis."},{"key":"ref_2","doi-asserted-by":"crossref","unstructured":"Petrelli, A., and Di Stefano, L. (2011, January 6\u201313). On the repeatability of the local reference frame for partial shape matching. Proceedings of the IEEE International Conference on Computer Vision, Barcelona, Spain.","DOI":"10.1109\/ICCV.2011.6126503"},{"key":"ref_3","doi-asserted-by":"crossref","unstructured":"Tombari, F., Salti, S., and Di Stefano, L. (2011, January 5\u201311). Unique signatures of histograms for local surface description. Proceedings of the European Conference on Computer Vision, Heraklion, Crete, Greece.","DOI":"10.1007\/978-3-642-15558-1_26"},{"key":"ref_4","doi-asserted-by":"crossref","unstructured":"Gojcic, Z., Zhou, C., Wegner, J.D., and Wieser, A. (2019, January 15\u201320). The perfect match: 3d point cloud matching with smoothed densities. Proceedings of the IEEE International Conference on Computer Vision and Pattern Recognition, Long Beach, CA, USA.","DOI":"10.1109\/CVPR.2019.00569"},{"key":"ref_5","unstructured":"Spezialetti, R., Salti, S., and Stefano, L.D. (November, January 27). Learning an Effective Equivariant 3D Descriptor Without Supervision. Proceedings of the IEEE International Conference on Computer Vision, Seoul, Korea."},{"key":"ref_6","first-page":"2","article-title":"Overview of the RANSAC Algorithm","volume":"4","author":"Derpanis","year":"2010","journal-title":"Image Rochester NY"},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Deng, H., Birdal, T., and Ilic, S. (2019). 3D Local Features for Direct Pairwise Registration. arXiv.","DOI":"10.1109\/CVPR.2019.00336"},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"3766","DOI":"10.1109\/TIP.2018.2827330","article-title":"Toward the Repeatability and Robustness of the Local Reference Frame for 3D Shape Matching: An Evaluation","volume":"27","author":"Yang","year":"2018","journal-title":"IEEE Trans. Image Proces."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"348","DOI":"10.1007\/s11263-009-0296-z","article-title":"On the repeatability and quality of keypoints for local feature-based 3d object retrieval from cluttered scenes","volume":"89","author":"Mian","year":"2010","journal-title":"Int. J. Comput. Vis."},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Petrelli, A., and Di Stefano, L. (2012, January 13\u201315). A repeatable and efficient canonical reference for surface matching. Proceedings of the Second International Conference on 3D Imaging, Modeling, Processing, Visualization & Transmission, Zurich, Switzerland.","DOI":"10.1109\/3DIMPVT.2012.51"},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"175","DOI":"10.1016\/j.patcog.2016.11.019","article-title":"TOLDI: An effective and robust approach for 3D local shape description","volume":"65","author":"Yang","year":"2017","journal-title":"Pattern Recognit."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"135","DOI":"10.1002\/cem.1122","article-title":"Resolving the sign ambiguity in the singular value decomposition","volume":"22","author":"Bro","year":"2008","journal-title":"J. Chemom."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"63","DOI":"10.1023\/A:1007981719186","article-title":"Point signatures: A new representation for 3d object recognition","volume":"25","author":"Chua","year":"1997","journal-title":"Int. J. Comput. Vis."},{"key":"ref_14","doi-asserted-by":"crossref","unstructured":"Deng, H., Birdal, T., and Ilic, S. (2018, January 8\u201314). Ppf-foldnet: Unsupervised learning of rotation invariant 3d local descriptors. Proceedings of the European Conference on Computer Vision, Munich, Germany.","DOI":"10.1007\/978-3-030-01228-1_37"},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"198","DOI":"10.1007\/s11263-012-0545-4","article-title":"Performance evaluation of 3D keypoint detectors","volume":"102","author":"Tombari","year":"2013","journal-title":"Int. J. Comput. Vis."},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"19","DOI":"10.1007\/s11263-005-3221-0","article-title":"A novel representation and feature matching algorithm for automatic pairwise registration of range images","volume":"66","author":"Mian","year":"2006","journal-title":"Int. J. Comput. Vis."},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"1584","DOI":"10.1109\/TPAMI.2006.213","article-title":"Three-dimensional model-based object recognition and segmentation in cluttered scenes","volume":"28","author":"Mian","year":"2006","journal-title":"IEEE Trans. Pattern Anal. Mach. Intel."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"1285","DOI":"10.1109\/TPAMI.2007.1060","article-title":"Snapshots: A novel local surface descriptor and matching algorithm for robust 3D surface alignment","volume":"29","author":"Malassiotis","year":"2007","journal-title":"IEEE Trans. Pattern Anal. Mach. Intel."},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"381","DOI":"10.1145\/358669.358692","article-title":"Random sample consensus: A paradigm for model fitting with applications to image analysis and automated cartography","volume":"24","author":"Fischler","year":"1981","journal-title":"Commun. ACM"},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"66","DOI":"10.1007\/s11263-015-0824-y","article-title":"A comprehensive performance evaluation of 3D local feature descriptors","volume":"116","author":"Guo","year":"2016","journal-title":"Int. J. Comput. Vis."},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Rusu, R.B., and Cousins, S. (2011, January 9\u201313). 3d is here: Point cloud library (pcl). Proceedings of the IEEE International Conference on Robotics and Automation, Shanghai, China.","DOI":"10.1109\/ICRA.2011.5980567"}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/20\/18\/5086\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,11]],"date-time":"2025-10-11T10:07:38Z","timestamp":1760177258000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/20\/18\/5086"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2020,9,7]]},"references-count":21,"journal-issue":{"issue":"18","published-online":{"date-parts":[[2020,9]]}},"alternative-id":["s20185086"],"URL":"https:\/\/doi.org\/10.3390\/s20185086","relation":{},"ISSN":["1424-8220"],"issn-type":[{"value":"1424-8220","type":"electronic"}],"subject":[],"published":{"date-parts":[[2020,9,7]]}}}