Closed Wei-i closed 2 years ago
oh! Thanks a lot! But this makes me strange because PCB module executes during inference
.
When I read your paper, I thought it may be a common module like 'channel wise add' or 'reweighting' during training
.
My question is that when we reweight the score
using PCB module, why can we improve AP?
I think during inference, for instance the input is a image while the output is the list of proposals, each proposal has score
, pred label
and pred box
. I am confused that I think coco eval api does not care about the scores
for a proposal because label
and bbox
is certain.
What is wrong with my view?
Thanks you :)
Hi, I am sorry that I can not find the code where the
PCB
module for re-weighting the box-predictor' cls scores refers to. Hope someone who knows could help me :), thx.