{"id":18792503,"url":"https://github.com/prbonn/auto-mos","last_synced_at":"2025-07-31T22:04:46.002Z","repository":{"id":55156154,"uuid":"465227392","full_name":"PRBonn/auto-mos","owner":"PRBonn","description":"Automatic Labeling to Generate Training Data for Online LiDAR-based Moving Object Segmentation","archived":false,"fork":false,"pushed_at":"2022-08-13T19:06:46.000Z","size":1260,"stargazers_count":79,"open_issues_count":2,"forks_count":5,"subscribers_count":10,"default_branch":"main","last_synced_at":"2023-11-17T04:03:13.857Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Python","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/PRBonn.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}},"created_at":"2022-03-02T08:54:18.000Z","updated_at":"2023-10-17T12:04:43.000Z","dependencies_parsed_at":"2022-08-14T13:41:05.779Z","dependency_job_id":null,"html_url":"https://github.com/PRBonn/auto-mos","commit_stats":null,"previous_names":[],"tags_count":0,"template":null,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/PRBonn%2Fauto-mos","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/PRBonn%2Fauto-mos/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/PRBonn%2Fauto-mos/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/PRBonn%2Fauto-mos/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/PRBonn","download_url":"https://codeload.github.com/PRBonn/auto-mos/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":223589790,"owners_count":17170037,"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":"2024-11-07T21:20:10.666Z","updated_at":"2024-11-07T21:20:11.224Z","avatar_url":"https://github.com/PRBonn.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Auto-MOS: Automatic Labeling to Generate Training Data for Online LiDAR-based Moving Object Segmentation\n\nThis repo contains the code for our Auto-MOS, which automatically generates training data for LiDAR-based moving objects segmentation [PDF](http://arxiv.org/pdf/2201.04501).\n\n\u003cimg src=\"pics/framework.png\" width=\"800\"\u003e\n\n### Table of Contents\n1. [Introduction](#Auto-MOS:-Automatic-Labeling-to-Generate-Training-Data-for-Online-LiDAR-based-Moving-Object-Segmentation)\n2. [Publication](#Publication)\n3. [Logs](#Logs)\n4. [Dependencies](#Dependencies)\n5. [How to use](#How-to-use)\n6. [Application](#Application)\n7. [License](#License)\n\n## Publication\nIf you use our implementation in your academic work, please cite the corresponding paper ([PDF](http://arxiv.org/pdf/2201.04501)):\n    \n\t@article{chen2022ral,\n\t         author      = {X. Chen and B. Mersch and L. Nunes and R. Marcuzzi and I. Vizzo and J. Behley and C. Stachniss},\n\t         title       = {{Automatic Labeling to Generate Training Data for Online LiDAR-Based Moving Object Segmentation}},\n\t         journal     = {IEEE Robotics and Automation Letters (RA-L)},\n\t         year        = 2022,\n\t         volume      = 7,\n\t         number      = 3,\n\t         pages       = {6107-6114},\n\t         url         = {http://arxiv.org/pdf/2201.04501},\n\t         issn        = {2377-3766},\n\t         doi         = {10.1109/LRA.2022.3166544}\n\t        }\n\t     \n## Logs\n\n### Version 1.0\n\nNote that, due to copyright and protection of our benchmark, this repo currently only provides the tracking and label generating parts of the proposed method.\nFor Odometry/LiDAR-SLAM we refer to our SuMa ([link](https://github.com/jbehley/SuMa)), refer dynamic removal to ERASOR ([link](https://github.com/LimHyungTae/ERASOR)), refer instance clustering to HDBSCAN ([link](https://github.com/scikit-learn-contrib/hdbscan)), and refer the LiDAR-MOS network to our LMNet ([link](https://github.com/PRBonn/LiDAR-MOS)).\n\t        \n## Dependencies\nBefore using our code, you need to install some libraries.\n\n- System dependencies:\n\n  ```bash\n  sudo apt-get update \n  sudo apt-get install -y python3-pip wget unzip\n  sudo -H pip3 install --upgrade pip\n  ```\n\n- Python dependencies (may also work with different versions than mentioned in the requirements file)\n\n  ```bash\n  sudo -H pip3 install -r requirements.txt\n  ```\n\t        \n## How to run\n\n### Download data and intermediate results\n\nTo run the quick demo, please first download the data ([link](https://www.ipb.uni-bonn.de/html/projects/auto-mos/kitti.zip)) extracting it to the `data` folder, and the intermediate instance results ([link](https://www.ipb.uni-bonn.de/html/projects/auto-mos/instances.zip)) extracting it to the `results` folder.\n\nTo visualize the final results, you could also directly download the mos results ([link](https://www.ipb.uni-bonn.de/html/projects/auto-mos/mos_predictions.zip)) and extract it into the `results` folder. \n\nYou could also download the data and intermediate results using command lines as follows:\n\n- Download kitti demo dataset:\n\n  ```bash\n  wget -P data/ https://www.ipb.uni-bonn.de/html/projects/auto-mos/kitti.zip\n  unzip data/kitti.zip -d data\n  rm data/kitti.zip\n  ```\n- Download instance predictions:\n\n  ```bash\n  wget -P results/ https://www.ipb.uni-bonn.de/html/projects/auto-mos/instances.zip\n  unzip results/instances.zip -d results\n  rm results/instances.zip\n  ```\n- Download final mos predictions:\n\n  ```bash\n  wget -P results/ https://www.ipb.uni-bonn.de/html/projects/auto-mos/mos_predictions.zip\n  unzip results/mos_predictions.zip -d results\n  rm results/mos_predictions.zip\n  ```\n  \t\n### Quick run\n\n- To automatic generate the mos labels, one could directly run:\n  ```bash\n  python3 auto-mos-tracking.py\n  ```\n  \n- To visualize the mos results, one could directly run:\n  ```bash\n  python3 vis_mos_results.py\n  ```\n  \n  To control the visualizer:\n  - press `n`: play next scan,\n  - press `b`: play previous scan,\n  - press `esc` or `q`: exits.\n   \n- To visualize the intermediate instance predictions, one could directly run:\n  ```bash\n  python3 vis_instances.py\n  ```\n  \n  To control the visualizer:\n  - press `esc` or `q`: exits.\n            \n## License\nThis project is free software made available under the MIT License. For details see the LICENSE file.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fprbonn%2Fauto-mos","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fprbonn%2Fauto-mos","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fprbonn%2Fauto-mos/lists"}