Marker-based SLAM

此项目是SPM-SLAM的改进版本，也是一种基于AprilTag或者Aruco类型标记（marker）的SLAM系统，通过在环境中布置不同ID的marker，即可快速实现高精度的相机定位，经过在实验室环境下的测试，可以达到厘米级精度。具体流程可看图1。绿色部分是对SPM-SLAM的修改部分。

Papers

[SPM-SLAM] Rafael Muñoz-Salinas and Manuel J. Marín-Jimenez and R. Medina-Carnicer, "SPM-SLAM: Simultaneous Localization and Mapping with quared Planar Markers", Pattern Recognition, Vol. 86, pp. 156-171, 2019.

安装需求

OpenCV

Required at least 3.4, Tested with OpenCV3.4.12

AprilTag 3

Download and instructions can be found at: AprilTag3 library.

Eigen3

Required by g2o. Required at least 3.1.1.

Install

mkdir build && cd build
cmake..
make -j4

Usage Example

执行：./spm-slam "video" "file1.yml" -conf "file2.yml" --outlog "cameraPose.txt"
video: 数据集
file1.yml: camera parameters
file2.yml: system parameters cameraPose.txt: output camera poses.