https://github.com/cleinc/slslam

Building a 3D Line-based Map Using a Stereo SLAM
https://github.com/cleinc/slslam

Last synced: 7 days ago
JSON representation

Building a 3D Line-based Map Using a Stereo SLAM

• Host: GitHub
• URL: https://github.com/cleinc/slslam
• Owner: cleinc
• License: gpl-2.0
• Created: 2015-03-23T10:56:16.000Z (over 9 years ago)
• Default Branch: master
• Last Pushed: 2022-11-02T12:50:56.000Z (about 2 years ago)
• Last Synced: 2023-11-07T15:24:13.323Z (about 1 year ago)
• Language: C++
• Size: 840 KB
• Stars: 108
• Watchers: 7
• Forks: 43
• Open Issues: 1
• Metadata Files:
  • Readme: README
  • License: LICENSE

Awesome Lists containing this project

• awesome-slam - CODE

README

        
*** SLSLAM ***

* Guoxuan Zhang, * Jin Han Lee, Jongwoo Lim, Il Hong Suh

* Equal contribution

This code implements a Stereo Line-based SLAM (SLSLAM) presented in

"Building a 3D Line-based Map Using a Stereo SLAM," IEEE Transactions on Robotics.

Dependencies: Ceres-solver 1.7.0, OpenCV 2.4.10, Google Flags 2.1.0, Google Logging 0.0.0, OpenGL, GLFW

All dependencies should be installed in your system beforehand.

This code has been tested only on 64-bit linux (Ubuntu 12.04 and 14.04) systems.

First download the code:

$ git clone https://github.com/slslam/slslam.git

$ cd slslam

Make a directory for datasets

$ mkdir data 

Then download a zipped dataset file from:

https://drive.google.com/file/d/0B3bB8rHbc3fWNGQ3YnhLczIwNG8/view?usp=sharing

and copy it to ${source_root_dir}/data.

$ cd data

$ unzip data.zip

$ cd ..

$ mkdir build && cd build

$ cmake ..

$ make

$ ./slslam_it3f

$ ./slslam_olympic4f

$ ./slslam_myungdong

* Currently, line detection, tracking, and loop closure detection modules are

not included due to license problem with a project supported by Korean

government.