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
Dear author, thank you for your outstanding work. I noticed that the paper mentioned “ Note that we do not use any classification prediction of the pre-trained 2D detectors by default as our main focus is enforcing the 3D detector to learn
localizing from the pre-trained 2D detector at this stage. ” But I found that in the code, the pseudo box category is still passed in as the gt category. May I ask what the specific situation is? I am looking forward to your reply.
The text was updated successfully, but these errors were encountered:
Dear author, thank you for your outstanding work. I noticed that the paper mentioned “ Note that we do not use any classification prediction of the pre-trained 2D detectors by default as our main focus is enforcing the 3D detector to learn
localizing from the pre-trained 2D detector at this stage. ” But I found that in the code, the pseudo box category is still passed in as the gt category. May I ask what the specific situation is? I am looking forward to your reply.
The text was updated successfully, but these errors were encountered: