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
I am trying to run msckf on WHUVID dataset with stereo+IMU, when running on the fast-speed dataset, there is drift in the beginning ,however then running on the low-speed dataset there is drift in the end.
The upper is fast-speed using evo tool to evaluate:
The other is low-speed:
And I wounder why is like this? I'd appreciate it for your response.
The text was updated successfully, but these errors were encountered:
And every time the total path length predicted by MSCKF is longer than the ground truth one. Is this because there is no stationary time in the beginning?
I am trying to run msckf on WHUVID dataset with stereo+IMU, when running on the fast-speed dataset, there is drift in the beginning ,however then running on the low-speed dataset there is drift in the end.
The upper is fast-speed using evo tool to evaluate:
The other is low-speed:
And I wounder why is like this? I'd appreciate it for your response.
The text was updated successfully, but these errors were encountered: