Giters

linxid / Focus-DETR-mindspore

[ICCV 2023] Official implementation of the paper "Less is More: Focus Attention for Efficient DETR"

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

What is the ground truth for training?

rginjapan opened this issue · comments

commented

Thanks for your work. When training the focal loss and multi-category score predictor, do you need the foreground and background label (ground truth)

commented

We use COCO label to generate foreground and background label. Pytorch code is avaliable in https://github.com/IDEA-Research/detrex/blob/main/projects/focus_detr/modeling/two_stage_criterion.py

commented

Thanks for your reply, about multi-category score predictor, does it mean select the top-k score in foreground token? "multi-category" is a little bit confused me.

commented

"Multi-category score predictor " is similar to topk token selector in the Encoder of DINO.