thodan/epos

Depth information in training

iroCl opened this issue · 2 comments

iroCl commented

Are depth images really necessary for training? I am quite confused because in your paper you say you only use RGB images, but when I try to train my own dataset, it requires depth images (when running calc_gt_info).

Hi! I am new in these stuff and had the same question regarding the depth images. What are they actually required for in training and testing? Was there any help about it?

Hello. The depth images are not used when training EPOS. All BOP datasets provide depth images though and some of the scripts in this repo require these images (mostly for historical reasons). The scripts should be easy to modify such as this requirement is removed.