search

PRBonn / lidar-bonnetal

Semantic and Instance Segmentation of LiDAR point clouds for autonomous driving
http://semantic-kitti.org
MIT License
915 stars 204 forks source link

How to create cali.txt and pose.txt data from custom data #113

Closed smileyr99 closed 3 months ago

smileyr99 commented 3 months ago

First of all, thank you for providing excellent code and models. I want to use RangeNet++ with my custom data. I understand that I need to format the data similar to semantic-kitti. So, I've labeled the data and converted it into .bin files, but I'm unsure how to create pose.txt and cali.txt files. For pose.txt, it seems like I need to use surfel-based SLAM approach (SuMa) to obtain global coordinate values, but as a beginner, I don't know how to do that. I would really appreciate it if you could help me. Thank you in advance.

jbehley commented 3 months ago

if you are just interested in semantic segmentation, you don't need poses or the calib.txt. RangeNet++ uses a single scan. Not sure what you want to achieve. Can you maybe provide more information on your concrete problem.

The poses and calib.txt follows the KITTI conventions: see https://github.com/jbehley/point_labeler/issues/23#issuecomment-640119577

For pose estimation on your own data, you can give kiss icp (see https://github.com/PRBonn/kiss-icp)