Fast-RCNN代码解读(0)
由于博主最近正在尝试修改rbg大神的fast-rcnn代码,为了推进自己的学习进度,在此做一个记录,同时也便于跟大家交流。
Fast-RCNN配置文件解读
root/lib/fast-rcnn/config.py参数解读
- cfg.DEDUP_BOXES
在Fast-rcnn中使用RoI Pooling将图像中的proposal坐标映射到feature map上时,由于feature map是很小的空间区域(Pool4之后feature map大小为 14×14 ),将会使得原本在原始图像上处于不同位置的proposal映射到feature map上的同一位置,从而产生不必要的冗余。所以设置参数deduplicated,其中设置值为1/16是由于在网络中经过4个Pooling( 2×2 ),feature map缩小了16倍。
# The mapping from image coordinates to feature map coordinates might cause
# some boxes that are distinct in image space to become identical in feature
# coordinates. If DEDUP_BOXES > 0, then DEDUP_BOXES is used as the scale factor
# for identifying duplicate boxes.
# 1/16 is correct for {Alex,Caffe}Net, VGG_CNN_M_1024, and VGG16
__C.DEDUP_BOXES = 1./16.