Distilling-Object-Detectors
Distilling-Object-Detectors copied to clipboard
when distillation with FPN,which feature map shuold we select to calculate imitation loss, feature map of FPN(P3、P4...) or feature of backbone(C3、C4...) or others?
Hi, fpn features (p3, p4...) are used.
when distilling Faster RCNN with FPN, do we need to firstly assign gt box to different FPN level and then generate imitation region? Or for every FPN level, we generate imitation region according to all gt boxes?
on each level of fpn, only the assigned gt box will be used calculate the imitation region
the question is how to assign gt box to different level? could you please explain in detail?