We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
As mentioned in this issue, ONCE dataset doesn't release camera extrinsics. How to train Persformer on ONCE and reproduce your latest results?