Faster RCNN consists of two modules.
(a) Region proposal network and (b) Fast R-CNN detector
The paper mention Region proposal network runs on the feature maps.
So, I have a input 256 x 256 image and I use VGG16 for example, after 5 times of pooling, the feature maps with only left size of 8 x 8.
My question is
How am I going to propose on the feature maps if the size is not the same as the original image?
Does the paper use any upsampling techniques to upsample the feature maps?