You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi,thanks for your great work.But some problem came to me.
There is self.use_icp = True in the coding , is it just mean using ICP to refine register model ? which called "Ours+ICP"method in paper.
However,we use KITTI dataset's sequence08 to test the "Ours+ICP" model, and the average RTE is 0.15760340035163767(m) ? and if we set self.use_icp = False, the average RTE is 0.37215860772969533(m) based on sequence08 .
Thanks for your great job again!
The text was updated successfully, but these errors were encountered:
Hi! I know why my answer was different from Tab3 now.Because in your file test_kitti.py ,there is succ_rate, rte, rre, avg_time, _ = stats[:i + 1].mean(0) so the logger will record all RTE and RRE including unsuccssed pairs.
Thanks!
Hi,thanks for your great work.But some problem came to me.
There is
self.use_icp = True
in the coding , is it just mean using ICP to refine register model ? which called "Ours+ICP"method in paper.However,we use KITTI dataset's sequence08 to test the "Ours+ICP" model, and the average RTE is 0.15760340035163767(m) ? and if we set
self.use_icp = False
, the average RTE is 0.37215860772969533(m) based on sequence08 .Thanks for your great job again!
The text was updated successfully, but these errors were encountered: