{"id":19559732,"url":"https://github.com/ShiPC-AI/TCF","last_synced_at":"2025-04-26T23:32:36.592Z","repository":{"id":255944014,"uuid":"852582002","full_name":"ShiPC-AI/TCF","owner":"ShiPC-AI","description":"A Two-stage Consensus Filtering for Real-time 3D Registration","archived":false,"fork":false,"pushed_at":"2024-09-08T01:21:05.000Z","size":5632,"stargazers_count":2,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2024-09-08T02:57:01.011Z","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":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/ShiPC-AI.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2024-09-05T03:59:15.000Z","updated_at":"2024-09-08T01:21:09.000Z","dependencies_parsed_at":null,"dependency_job_id":"8a20a0a1-dd4c-4261-a86c-41019e1668c6","html_url":"https://github.com/ShiPC-AI/TCF","commit_stats":null,"previous_names":["shipc-ai/tcf"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ShiPC-AI%2FTCF","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ShiPC-AI%2FTCF/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ShiPC-AI%2FTCF/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/ShiPC-AI%2FTCF/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/ShiPC-AI","download_url":"https://codeload.github.com/ShiPC-AI/TCF/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":251068040,"owners_count":21531475,"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","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":"2024-11-11T05:01:11.449Z","updated_at":"2025-04-26T23:32:36.585Z","avatar_url":"https://github.com/ShiPC-AI.png","language":"C++","funding_links":[],"categories":["C++"],"sub_categories":[],"readme":"\u003ch1 align=\"center\"\u003eTCF (Two-stage Consensus Filtering)\u003c/h1\u003e\n\n\u003cp style=\"text-align: justify;\"\u003e\n\u003cstrong\u003eTCF\u003c/strong\u003e is a 3D correspondence-based point cloud registration method. It elevates RANSAC to SOTA speed and accuracy. Firstly, one-point RANSAC obtains a consensus set based on length consistency. Subsequently, two-point RANSAC refines the set via angle consistency. Then, three-point RANSAC computes a coarse pose and removes outliers based on transformed correspondence’s distances. Drawing on optimizations from one-point and two-point RANSAC, three-point RANSAC requires only a few iterations. Eventually, an iterative reweighted least squares (IRLS) is applied to yield the optimal pose.\n\u003c/p\u003e\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"figures/framework-a.png\" alt=\"Subfigure 1\" width=\"85%\"\u003e\u003cbr\u003e\n  \u003cem\u003e(a) Overall framework\u003c/em\u003e\n\u003c/p\u003e\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"figures/framework-b.png\" alt=\"Subfigure 2\" width=\"85%\"\u003e\u003cbr\u003e\n  \u003cem\u003e(b) Outlier Removal\u003c/em\u003e\n\u003c/p\u003e\n\n##  News\n- **2025.01:** The preprocessed dataset in TCF is available: [Dataset Available](https://drive.google.com/drive/folders/1uKN2pqymFc85tY13Cw7kUNop7uQOEOSB).\n- **2024.11:** TCF has been accepted for *IEEE Robotics and Automation Letters*: [Published Paper](https://ieeexplore.ieee.org/document/10758239).\n\n##  Citation\nIf you use code or data of TCF in your academic research, please cite our paper:\n```\n@ARTICLE{10758239,\n  author={Shi, Pengcheng and Yan, Shaocheng and Xiao, Yilin and Liu, Xinyi and Zhang, Yongjun and Li, Jiayuan},\n  journal={IEEE Robotics and Automation Letters}, \n  title={RANSAC Back to SOTA: A Two-Stage Consensus Filtering for Real-Time 3D Registration}, \n  year={2024},\n  volume={9},\n  number={12},\n  pages={11881-11888},\n  doi={10.1109/LRA.2024.3502056}}\n```\n\n## Dependencies\n- **CMake**\n- **PCL** (Point Cloud Library)\n- **Eigen ≥ 3.4** (for slicing matrix)\n- **nlohmann_json** (for reading config files)\n\n## File and data structure\n### Configure files\n```bash\nconfig\n│   ├── config_eth.json\n│   └── config_kitti.json\n```\n### Data structure\n```bash\ndata\n├── ETH_TLS\n│   └── facade\n│       ├── s1_v0.1_sor.pcd\n│       ├── s2_v0.1_sor.pcd\n│       ├── s2_s1_top3.match\n│       ├── s2-s1.pose\n│       └── ...\n├── KITTI\n│   └── 09\n│       ├── 210_v0.3.pcd\n│       ├── 191_v0.3.pcd\n│       ├── 210_191_top3.match\n│       ├── 210-191.pose\n│       └── ...\n```\n- **xx.json**: Contains paths for source point cloud, target point cloud, initial correspondences, and ground truth pose.\n- **xx.pcd**: Source and target point cloud.\n- **A_B_top3.match**: A correspondence file where A is source (left 3 columns) and B is target (right 3 columns).\n- **xx.pose**: The 4*4 groundtruth pose from A to B\n\n**Note:** Source and target point clouds are used to calculate resolution and estimate noise level, enabling **a genralized method without manual parameter adjustments**. If not available, values can be set manually, but correspondences remain necessary.\n\n## How to use?\n### Complie and run Demo\n```bash\nmkdir build \u0026\u0026 cd build\ncmake ..\nmake -j8\n./demo\n```\n### Run other data\nTo test with different data, simply update the data path in the config file.","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FShiPC-AI%2FTCF","html_url":"https://awesome.ecosyste.ms/projects/github.com%2FShiPC-AI%2FTCF","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FShiPC-AI%2FTCF/lists"}