{"id":21195039,"url":"https://github.com/plusk01/pointcloud-plane-segmentation","last_synced_at":"2025-07-27T07:08:29.791Z","repository":{"id":55147919,"uuid":"301833868","full_name":"plusk01/pointcloud-plane-segmentation","owner":"plusk01","description":null,"archived":false,"fork":false,"pushed_at":"2021-01-07T02:11:41.000Z","size":917,"stargazers_count":72,"open_issues_count":0,"forks_count":16,"subscribers_count":3,"default_branch":"master","last_synced_at":"2025-07-10T08:08:46.630Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"C++","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/plusk01.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null}},"created_at":"2020-10-06T19:30:04.000Z","updated_at":"2025-05-23T11:07:38.000Z","dependencies_parsed_at":"2022-08-14T13:30:58.907Z","dependency_job_id":null,"html_url":"https://github.com/plusk01/pointcloud-plane-segmentation","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/plusk01/pointcloud-plane-segmentation","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/plusk01%2Fpointcloud-plane-segmentation","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/plusk01%2Fpointcloud-plane-segmentation/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/plusk01%2Fpointcloud-plane-segmentation/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/plusk01%2Fpointcloud-plane-segmentation/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/plusk01","download_url":"https://codeload.github.com/plusk01/pointcloud-plane-segmentation/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/plusk01%2Fpointcloud-plane-segmentation/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":267320257,"owners_count":24068527,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","status":"online","status_checked_at":"2025-07-27T02:00:11.917Z","response_time":82,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2024-11-20T19:25:46.975Z","updated_at":"2025-07-27T07:08:29.733Z","avatar_url":"https://github.com/plusk01.png","language":"C++","funding_links":[],"categories":[],"sub_categories":[],"readme":"Point Cloud Plane Segmentation\n==============================\n\nThis library can be used to extract planar patches from unorganized point clouds. It is a stripped-down version of [**@abnerrjo**](https://github.com/abnerrjo)'s Robust Statistics-based Plane Detection (`rspd`) work published as\n\n\u003e AMC Araújo and MM Oliveira, “A robust statistics approach for plane detection in unorganized point clouds,” Pattern Recognition, 2020.\n\nThe original implementation and other details can be found on [GitHub](https://github.com/abnerrjo/PlaneDetection) and their [project page](https://www.inf.ufrgs.br/~oliveira/pubs_files/RE/RE.html).\n\n## Look and Feel\n\nWhen executed on a raw point cloud with 50k points, `rspd` finds 20 planar patches in approximately 120 ms (i9-7920X, 64GB).\n\n\u003cp align=\"middle\"\u003e\n  \u003cimg src=\".github/bluepcd.png\" width=\"49%\" /\u003e\n  \u003cimg src=\".github/bluepcd-with-planes.png\" width=\"49%\" /\u003e\n\u003c/p\u003e\n\n## Build and Run\n\nThis library and an example app can be built with the usual\n\n```bash\n$ mkdir build \u0026\u0026 cd build\n$ cmake .. \u0026\u0026 make\n```\n\nUsing the provided test data, you can run `rspd` on a point cloud with\n\n```bash\n$ cd build\n$ ./main ../data/blue.pcd\n```\n\n### Open3D Dependency\n\nThe `master` branch depends on [Open3D](http://www.open3d.org/). If you did not install Open3D system-wide, you will need to provide the `CMAKE_PREFIX_PATH` instead of the `cmake ..` command above\n\n```bash\n$ cmake -DCMAKE_PREFIX_PATH=/path/to/Open3D/build/install ..\n```\n\nIn this case, **Open3D was built from source** after running the `cmake` command\n\n```bash\n# Open3D was built from source using\n$ cmake -DCMAKE_INSTALL_PREFIX=$(pwd)/install ..\n```\n\n## Parameters\n\n| Parameter                            | Default                                 | Description                                                                                                                                                                                                                                    |\n|--------------------------------------|-----------------------------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|\n| `mMinNormalDiff`                     | 60 degrees                              | A patch's associated point normals are tested against the patch's estimated normal. If the lower bound of the spread of similarity scores (roughly: 3σ below median) is lower than this, the patch is considered not robust and is thrown out. |\n| `mMaxDist`                           | 75 degrees                              | A patch's associated points are scored on distance to plane. Smaller values encourage tighter distribution of points around the plane. Also encourages truly planar patches. See Fig 4 of paper.                                               |\n| `mOutlierRatio`                      | 0.75                                    | Maximum allowable ratio of outlier points in a given patch's associated point set.                                                                                                                                                             |\n| `minNumPoints`                       | 30                                      | Do not subdivide an octree node with less than this number of points                                                                                                                                                                           |\n| `nrNeighbors` (normal estimation)    | 75                                      | Number of neighbors used to estimate the normals of each point.                                                                                                                                                                                |\n| `nrNeighbors` (knn search)           | 75                                      | Number of neighbors used when attempting to grow and merge patches. More neighbors will result in (potentially) fewer output patches at the expense of more runtime.                                                                           |\n| minimum length of longest patch edge | 0.01 * longest dimension of point cloud | In the original code, this parameter is hardcoded. This value is used to detect false positives and may be a valuable knob to tune. Alternatively, area could be used to cull too small planes.                                                |\n\n## Differences with Original Implementation\n\nThis implementation only provides the `rspd` functionality for planar patch detection. Additionally, it removes the Qt dependency and uses modern CMake for the build system. The algorithm and default parameters are the same as in [the original](https://github.com/abnerrjo/PlaneDetection).\n\nAlthough the `master` branch `rspd` library depends on Open3D, the [`rspd-original`](https://github.com/plusk01/pointcloud-plane-segmentation/tree/original-rspd) branch does not (the `rspd` library, though the app used to test / visualize does).\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fplusk01%2Fpointcloud-plane-segmentation","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fplusk01%2Fpointcloud-plane-segmentation","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fplusk01%2Fpointcloud-plane-segmentation/lists"}