{"id":16552301,"url":"https://github.com/cggos/lsd_slam_cg","last_synced_at":"2026-05-30T18:31:34.384Z","repository":{"id":108875396,"uuid":"186270663","full_name":"cggos/lsd_slam_cg","owner":"cggos","description":"Modified version of tum-vision/lsd_slam (commit d1e6f0e on Dec 12, 2014), Ubuntu 16.04, ROS Kinetic","archived":false,"fork":false,"pushed_at":"2019-11-17T08:57:17.000Z","size":1103,"stargazers_count":1,"open_issues_count":0,"forks_count":1,"subscribers_count":3,"default_branch":"master","last_synced_at":"2025-03-04T16:48:20.473Z","etag":null,"topics":["slam","visual-odometry"],"latest_commit_sha":null,"homepage":"","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/cggos.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,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2019-05-12T15:17:44.000Z","updated_at":"2019-11-17T08:58:03.000Z","dependencies_parsed_at":"2023-04-24T16:18:02.312Z","dependency_job_id":null,"html_url":"https://github.com/cggos/lsd_slam_cg","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/cggos/lsd_slam_cg","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/cggos%2Flsd_slam_cg","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/cggos%2Flsd_slam_cg/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/cggos%2Flsd_slam_cg/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/cggos%2Flsd_slam_cg/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/cggos","download_url":"https://codeload.github.com/cggos/lsd_slam_cg/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/cggos%2Flsd_slam_cg/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":33705207,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-26T15:22:16.424Z","status":"online","status_checked_at":"2026-05-30T02:00:06.278Z","response_time":92,"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":["slam","visual-odometry"],"created_at":"2024-10-11T19:44:31.076Z","updated_at":"2026-05-30T18:31:34.379Z","avatar_url":"https://github.com/cggos.png","language":"C++","funding_links":[],"categories":[],"sub_categories":[],"readme":"# lsd_slam_cg\n\nModified version of [tum-vision/lsd_slam](https://github.com/tum-vision/lsd_slam) (commit d1e6f0e on Dec 12, 2014)\n\n**LSD-SLAM (Large-Scale Direct Monocular SLAM)** is a novel approach to real-time monocular SLAM. It is fully direct (i.e. does not use keypoints / features) and creates large-scale, semi-dense maps in real-time on a laptop.\n\nPapers:\n  - *LSD-SLAM: Large-Scale Direct Monocular SLAM, J. Engel, T. Schöps, D. Cremers*, ECCV '14\n  - *Semi-Dense Visual Odometry for a Monocular Camera, J. Engel, J. Sturm, D. Cremers*, ICCV '13\n\nLink:\n  - http://vision.in.tum.de/lsdslam\n\n-----\n\n# Install \u0026 Build\n\n```sh\nsudo apt install libsuitesparse-dev libqglviewer-dev-qt4 ros-kinetic-libg2o\nsudo ln -s /usr/lib/x86_64-linux-gnu/libQGLViewer-qt4.so \\\n           /usr/lib/x86_64-linux-gnu/libQGLViewer.so  \n\nmkdir -p ws_lsd_slam/src \u0026 cd ws_lsd_slam/src\ngit clone https://github.com/cggos/lsd_slam_cg.git\ncd ..\ncatkin_make -j2\n```\n\n# Run\n\n* run with  [Room Example Sequence](http://vmcremers8.informatik.tu-muenchen.de/lsd/LSD_room.bag.zip)\n\n\t```sh\n\trosrun lsd_slam_viewer viewer\n\trosrun lsd_slam_core live_slam image:=/image_raw camera_info:=/camera_info\n\trosbag play ~/LSD_room.bag\n\t```\n\n  \u003cdiv align=center\u003e\n    \u003cimg src=\"lsd_slam_viewer.jpg\"\u003e\n  \u003c/div\u003e\n\n# Loop-Closure\n\n## openFabMap for large loop-closure detection\n\nIf you want to use openFABMAP for large loop closure detection, uncomment the following lines in `lsd_slam_core/CMakeLists.txt` :\n\n    #add_subdirectory(${PROJECT_SOURCE_DIR}/thirdparty/openFabMap)\n    #include_directories(${PROJECT_SOURCE_DIR}/thirdparty/openFabMap/include)\n    #add_definitions(\"-DHAVE_FABMAP\")\n    #set(FABMAP_LIB openFABMAP )\n\n**Note for Ubuntu 14.04:** The packaged OpenCV for Ubuntu 14.04 does not include the nonfree module, which is required for openFabMap (which requires SURF features).\nYou need to get a full version of OpenCV with nonfree module, which is easiest by compiling your own version.\nWe suggest to use the [2.4.8](https://github.com/Itseez/opencv/releases/tag/2.4.8) version, to assure compatibility with the current indigo open-cv package.\n\n\n# Troubleshoot / FAQ\n\n**How can I get the live-pointcloud in ROS to use with RVIZ?**\n\nYou cannot, at least not on-line and in real-time. The reason is the following:\n\nIn the background, LSD-SLAM continuously optimizes the pose-graph, i.e., the poses of all keyframes. Each time a keyframe's pose changes (which happens all the time, if only by a little bit), all points from this keyframe change their 3D position with it. Hence, you would have to continuously re-publish and re-compute the whole pointcloud (at 100k points per keyframe and up to 1000 keyframes for the longer sequences, that's 100 million points, i.e., ~1.6GB), which would crush real-time performance.\n\nInstead, this is solved in LSD-SLAM by publishing keyframes and their poses separately:\n- keyframeGraphMsg contains the updated pose of each keyframe, nothing else.\n- keyframeMsg contains one frame with it's pose, and - if it is a keyframe - it's points in the form of a depth map.\n\nPoints are then always kept in their keyframe's coodinate system: That way, a keyframe's pose can be changed without even touching the points. In fact, in the viewer, the points in the keyframe's coodinate frame are moved to a GLBuffer immediately and never touched again - the only thing that changes is the pushed modelViewMatrix before rendering.\n\nNote that \"pose\" always refers to a Sim3 pose (7DoF, including scale) - which ROS doesn't even have a message type for.\n\nIf you need some other way in which the map is published (e.g. publish the whole pointcloud as ROS standard message as a service), the easiest is to implement your own Output3DWrapper.\n\n\n**Tracking immediately diverges / I keep getting \"TRACKING LOST for frame 34 (0.00% good Points, which is -nan% of available points, DIVERGED)!\"**\n- double-check your camera calibration.\n- try more translational movement and less roational movement\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fcggos%2Flsd_slam_cg","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fcggos%2Flsd_slam_cg","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fcggos%2Flsd_slam_cg/lists"}