{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/mining-point-cloud-local-structures-by-kernel","title":"Mining Point Cloud Local Structures by Kernel Correlation and Graph Pooling","arxiv_id":"1712.06760","date":"2017-12-19","proceeding":"CVPR 2018 6","authors":["Yiru Shen","Chen Feng","Yaoqing Yang","Dong Tian"],"abstract":"Unlike on images, semantic learning on 3D point clouds using a deep network\nis challenging due to the naturally unordered data structure. Among existing\nworks, PointNet has achieved promising results by directly learning on point\nsets. However, it does not take full advantage of a point's local neighborhood\nthat contains fine-grained structural information which turns out to be helpful\ntowards better semantic learning. In this regard, we present two new operations\nto improve PointNet with a more efficient exploitation of local structures. The\nfirst one focuses on local 3D geometric structures. In analogy to a convolution\nkernel for images, we define a point-set kernel as a set of learnable 3D points\nthat jointly respond to a set of neighboring data points according to their\ngeometric affinities measured by kernel correlation, adapted from a similar\ntechnique for point cloud registration. The second one exploits local\nhigh-dimensional feature structures by recursive feature aggregation on a\nnearest-neighbor-graph computed from 3D positions. Experiments show that our\nnetwork can efficiently capture local information and robustly achieve better\nperformances on major datasets. Our code is available at\nhttp://www.merl.com/research/license#KCNet","url_abs":"http://arxiv.org/abs/1712.06760v2","url_pdf":"http://arxiv.org/pdf/1712.06760v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"mining-point-cloud-local-structures-by-kernel","repo_url":"https://github.com/ftdlyc/KCNet_Pytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"point-cloud-registration","task_name":"Point Cloud Registration"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1712.06760","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}