wh200720041 / floam

Fast LOAM: Fast and Optimized Lidar Odometry And Mapping for indoor/outdoor localization IROS 2021
Other
1.07k stars 278 forks source link

How to reproduce the results of your Kitti ranking #22

Open xuwuzhou opened 3 years ago

xuwuzhou commented 3 years ago

Thank you for your great work!

   But I wonder that, What changes have you made to make your results on KITTI rank is 16th with translation error only 0.73%. Because the result which I get from your open-source code is much worse than 0.73%, which is about 2%.
  What parameters should I change to get your KITTI results?

  Best wishes!
  Sincerely~
juliangaal commented 1 year ago

How did you convert the ROS poses to KITTI compatible ones?

jorgeffonte commented 4 months ago

How did you convert the ROS poses to KITTI compatible ones?

In the KITTI calibration files, you have the transformation between the LiDAR frame (to which the ROS poses are referenced) and the camera frame (to which the KITTI GT poses are referenced). Create a script to transform each pose in the path.