kandi background

Landmarks_vSLAM | Landmarkbased Visual SLAM The idea behind

Download this library from

kandi X-RAY | Landmarks_vSLAM Summary

Landmarks_vSLAM is a Python library. Landmarks_vSLAM has no vulnerabilities, it has build file available, it has a Strong Copyleft License and it has low support. However Landmarks_vSLAM has 4 bugs. You can download it from GitHub.
Landmark-based Visual SLAM 🤖🤖🤖 The idea behind the project is to perform landmark-based visual SLAM detecting static objects as landmarks with object detection Currently the system is not 100% V-SLAM, it uses LIDAR scans too and detects landmarks with YOLO - still awesome though 😄. Landmarks are detected with You-Only-Look-Once - YOLO object detection Graph SLAM by cartographer-project. Cool videos of the results https://www.youtube.com/watch?v=k3LRF8AGJbI&t=49s https://www.youtube.com/watch?v=6s2ePNo1SqA&t=27s. Completed as a master thesis at Technical University of Denmark in collaboration with Mobile Industrial Robots Systems that we use: https://github.com/ultralytics/yolov5 https://github.com/cartographer-project/cartographer Robot provided by Mobile Industrial Robots - MiR100 https://www.mobile-industrial-robots.com/da/solutions/robots/mir100/ You also need RGB-D camera and ROS (we used ROS Melodic). 3D models used as landmarks found at: https://app.ignitionrobotics.org/OpenRobotics. Attempt for a setting up and use tutorial 😅😅😅. Before running it please become familiar with cartographer's ROS integration: https://google-cartographer-ros.readthedocs.io/en/latest/index.html.

kandi-support Support

  • Landmarks_vSLAM has a low active ecosystem.
  • It has 1 star(s) with 1 fork(s). There are 1 watchers for this library.
  • It had no major release in the last 12 months.
  • Landmarks_vSLAM has no issues reported. There are no pull requests.
  • It has a neutral sentiment in the developer community.
  • The latest version of Landmarks_vSLAM is current.

quality kandi Quality

  • Landmarks_vSLAM has 4 bugs (3 blocker, 0 critical, 0 major, 1 minor) and 210 code smells.


  • Landmarks_vSLAM has no vulnerabilities reported, and its dependent libraries have no vulnerabilities reported.
  • Landmarks_vSLAM code analysis shows 0 unresolved vulnerabilities.
  • There are 21 security hotspots that need review.

license License

  • Landmarks_vSLAM is licensed under the GPL-3.0 License. This license is Strong Copyleft.
  • Strong Copyleft licenses enforce sharing, and you can use them when creating open source projects.


  • Landmarks_vSLAM releases are not available. You will need to build from source code and install.
  • Build file is available. You can build the component from source.
  • It has 4243 lines of code, 222 functions and 28 files.
  • It has high code complexity. Code complexity directly impacts maintainability of the code.
Top functions reviewed by kandi - BETA

kandi has reviewed Landmarks_vSLAM and discovered the below as its top functions. This is intended to give you an instant insight into Landmarks_vSLAM implemented functionality, and help decide if they suit your requirements.

  • Train the model .
  • Initialize the model .
  • Runs the detection on the GPU .
  • Calculate the k mean anchors
  • Runs a prediction on a prediction .
  • Generate image for images .
  • Generate new landmark objects for a given image .
  • Generate random affine
  • Compute the loss for a given model .
  • Compute the precision recall curve .

Landmarks_vSLAM Key Features

Anastasia Panaretou https://github.com/anastasiapan

Phillip Mastrup https://github.com/PMastrup

First download and set-up ROS http://wiki.ros.org/ROS/Installation

Then setup Cartographer System https://google-cartographer-ros.readthedocs.io/en/latest/compilation.html

Cartographer system modifications We did not create a package for this. You can just merge the files under 'cartographer_ros' in the directories they appear to be. They should be under the path: catkin_ws/src/cartographer_ros/cartographer_ros Only one source file has been slightly modified to visualise landmarks with 3D models - completely optional

Setup YOLO by ultralytics https://github.com/ultralytics/yolov5 We trained for our own objects - currently traffic cones and fire extinguishers If you don't feel like training, COCO can be used

Find your camera drivers: We used astra from Orbecc https://astra-wiki.readthedocs.io/en/latest/installation.html

Since an RGB-D camera is used you need to setup OpenNI library from pip pip install openni pip install primesense

This method is based on Bag-of-Visual-Words and SURF If not familiar with this method watch this amazing video: https://www.youtube.com/watch?v=a4cFONdc6nc and check out this repo https://github.com/ovysotska/in_simple_english/blob/master/bag_of_visual_words.ipynb Whatever vocabulary can be used (based on SURF) OR create your own (which we did) following the instructions in codebook_creation/create_codebook.py

Start the detector with python : python detect_landmarks_vslam.py

Launch cartographer : roslaunch cartographer_ros vslam_2D_landmarks.launch

If you want to test on a bagfile : roslaunch cartographer_ros landmarks_2D.launch bag_filename:=path/to/your/bagfile.bag

Landmarks_vSLAM Examples and Code Snippets

No Code Snippets are available at this moment for Landmarks_vSLAM.Refer to component home page for details.

No Code Snippets are available at this moment for Landmarks_vSLAM.Refer to component home page for details.

Community Discussions

No Community Discussions are available at this moment for Landmarks_vSLAM.Refer to stack overflow page for discussions.

No Community Discussions are available at this moment for Landmarks_vSLAM.Refer to stack overflow page for discussions.

Community Discussions, Code Snippets contain sources that include Stack Exchange Network


No vulnerabilities reported

Install Landmarks_vSLAM

You can download it from GitHub.
You can use Landmarks_vSLAM like any standard Python library. You will need to make sure that you have a development environment consisting of a Python distribution including header files, a compiler, pip, and git installed. Make sure that your pip, setuptools, and wheel are up to date. When using pip it is generally recommended to install packages in a virtual environment to avoid changes to the system.


For any new features, suggestions and bugs create an issue on GitHub. If you have any questions check and ask questions on community page Stack Overflow .