Trying to understand this paper for LP recognition.
Three ROI Pooling layers are used to have features for 7 classifiers.
First one is for Province, second one is for alphabets and the rest are for five alphanumeric characters. The plate sample is Chinese LP.
I am confused how these seven classifiers know which respective part of the plate to look at for recognition.
Because all seven classifiers are fed with same concatenated three ROIPooling features as follows.
roi1 = roi_pooling_ims(_x1, boxNew.mm(p1), size=(16, 8))
roi2 = roi_pooling_ims(_x3, boxNew.mm(p2), size=(16, 8))
roi3 = roi_pooling_ims(_x5, boxNew.mm(p3), size=(16, 8))
rois = torch.cat((roi1, roi2, roi3), 1)
_rois = rois.view(rois.size(0), -1)
y0 = self.classifier1(_rois)
y1 = self.classifier2(_rois)
y2 = self.classifier3(_rois)
y3 = self.classifier4(_rois)
y4 = self.classifier5(_rois)
y5 = self.classifier6(_rois)
y6 = self.classifier7(_rois)