Junda24/AFNet

KITTI data split.

Opened this issue · 1 comments

When evaling, I found that in the split file of the kitti dataset, the val part uses the train dataset, and the train part also uses the val data. Why is this?

When evaling, I found that in the split file of the kitti dataset, the val part uses the train dataset, and the train part also uses the val data. Why is this?

This constitutes the standard kitti eigen split.