kaist-help

A tool to convert KAIST urban dataset to rosbag, and recover the ring and time filed of LiDAR pointcloud for LIO-SAM / LVI-SAM using.

---

How to use

1. Download desired database files from the website.

2. Extract .gz.tar files into their folders.

   find . -name 'urban28*.tar.gz' -execdir tar -xzvf '{}' \;
   

3. Make ros workspace and build.

   mkdir -p ~/kaist-help-ws/src
   cp -r kaist-help ~/kaist-help-ws/src
   cd ~/kaist-help-ws
   catkin build
   

4. Edit the kaist2bag/config/config.yaml with path.

   # directory of the raw KAIST dataset
   dataset: "/mnt/F/DataSet/KAIST_Urban_dataset/urban26/urban26-dongtan"
   # directory of the rosbag file to save
   save_to: "/mnt/F/DataSet/KAIST_Urban_dataset/urban26/urban26-dongtan"
   

5. Create a rosbag file for each sensor type.

   source devel/setup.bash
   roslaunch kaist2bag kaist2bag.launch
   

6. Merge all bags into a single one (if desired).

   rosrun kaist2bag mergebag.py merged.bag <bag_file_1> ... <bag_file_8>
   

---

Acknowledge

• The main package is from kaist2bag.
• The code of recovering ring and time filed is adapted from clins.