In the previous article, we have installed and run ORB_SLAM2, downloaded and compiled (including error reporting) in the article:
And we ran it using the EuRoC dataset. Today we use the framework to run the KITTI data set.
Note : If the EuRoC data set is not successfully run, this tutorial may not be suitable, because when the EuRoC data set is successfully run, ORB_SLAM2 has been successfully compiled. After the ros space is created, directly copy the entire compiled folder into it. Then catkin_make, etc., and then run the data set.
But if you are starting from scratch and have not compiled ORBSLAM2 but just downloaded it, you can consider trying this tutorial:
ORB-SLAM2 compiles, installs and runs KITTI data set_Jinterest's blog-CSDN blog_orb-slam runs kitti
I haven't run this yet, but this one is more comprehensive and I might give it a try.
Table of contents
1. Download the KITTI data set
2. Compile ORB_SLAM2 with ROS and run it
1. Download the KITTI data set
The official website of the data set is:
The KITTI Vision Benchmark Suite
According to ORBSLAM2, we just need to download the grayscale version of the data set, which is the first one.
This may need to be downloaded over the wall. If you can't download it, you can try over the wall, or find someone else to download it. 22GB downloads are also relatively slow.
After downloading, extract it into the dataset folder:
2. Compile ORB_SLAM2 with ROS and run it
1. Create ros space
Create a ros space in the selected folder
mkdir -p orb2-catkin_ws/srcAfterwards, you need to put the source code in the src folder. If the EuRoC data set has not been successfully run, this tutorial may not be suitable. After downloading the source code, you need to compile it successfully.
According to the previous tutorial, if the EuRoC data set is successfully run, just copy the entire ORB_SLAM2 folder into the src folder.
After that, in the orb2-catkin_ws folder, continue the compilation of ros space, etc.
cd orb2-catkin_ws
catkin_make -j8
source devel/setup.bashIn this way, the ros space is built, and we can run the data set according to the commands on the official website.
2. Run the dataset
cd src
./Examples/Monocular/mono_kitti Vocabulary/ORBvoc.txt Examples/Monocular/KITTI00-02.yaml 数据集路径/dataset/sequences/00/
Among them, dataset is the name of the folder after decompressing the KITTI dataset we downloaded.
3. Operation results
