Question about selecting DAM as ground truth of salient token prediction

kakaobrain / sparse-detr

PyTorch Implementation of Sparse DETR

Apache License 2.0

162 stars 16 forks source link

Open konyul opened 1 year ago

konyul commented 1 year ago

why didn't you set the real ground truth as ground truth of salient token prediction??

suojinhui commented 9 months ago

The ground truth is a box, unless you use a semantically segmented mask.