graspnet/graspnet-baseline

About load_grasp_labels and objectness_label in graspnet_dataset.py

rhett-chen opened this issue · 1 comments

rhett-chen commented

Hi，thanks for your great work！I have two questions about your code：

  1. graspnet-baseline/dataset/graspnet_dataset.py

    Line 199 in df67956

    objectness_label[objectness_label>1] = 1

    You set objectness_label > 1, which will lose the label of object id=0 in segmentation mask. I want to know why it's not objectness_label > 0?

  2. graspnet-baseline/dataset/graspnet_dataset.py

    Line 255 in df67956

    if i == 18: continue

    I wonder why you just skip the object id=18 in load_grasp_labels function？
    Hope to hear your response soon, thanks!
chenxi-wang commented

Hi, the original objectness label contains object ids from 0 to n, where n is the number of objects and object 0 indicates background. It shares the same format with YCB-Video dataset. Hence, objectness>1 and objectness>0 bring the same results.
The grasp label of object 18 is too large, which leads to OOM. So I skipped this object in training.