How does Vkitti get the amount of data?

[Mingyang-Du]

Why is the data volume of 33k downloaded from the four scenarios corresponding to vkitti, and how is the data volume of 20k in the paper obtained?

[Mingyang-Du]

We fully follow the vkitti dataset process, but 02', '06', '18', '20' are 33k, not 20k