hku-mars/STD

Ground-Truth of KA Urban East Dataset

PeiliSong501 opened this issue · 4 comments

Thank you for sharing your newest research! I've been following your works for a while, and I really admire your previous achievements.
I noticed that KA Urban East, the dataset used for testing your algorithm on Livox Horizon LiDAR, doesn't contain the ground truth poses according to its site(https://isas-server.iar.kit.edu/lidardataset/). It maybe silly to ask, but how did you generate the gt data?
Looking forward to your reply, and hope your paper be accepted soon!

We use the loop detection module and the graph optimization module to get a globally consistent map and calculate the gt value of the loop based on this.

Thanks, do you think different algorithms(loop detection, odometry, etc.) applied make obvious differences on the calculated gt result?

Thanks, do you think different algorithms(loop detection, odometry, etc.) applied make obvious differences on the calculated gt result?

If you can get a consistent global map through your algorithm, I don't think there is a significant difference between different algorithms for gt calculation.

Thank you very much for your patient reply!